International Journal of Computer Applications in Technology (87 papers in press)

Regular Issues

• PUID: persona user intention detection  
  by Shengjia Cui, Xianglong Qi, Xiao Wang, Chen Zeng 
  Abstract: Although user intention detection has been widely studied, existing researches suffer inferior performance when only using the semantic features of query and neglecting the personalized user attributes. A key challenge is that the same or similar queries among users who possess different social positions can be inferred from different intentions. Therefore, we propose a novel task that user attributes are introduced as additional personalised features in user intention detection besides the semantic information of queries, named Persona User Intention Detection (PUID). We collected the query log with corresponding user attributes on the professional search engine to construct a large-scale user intention dataset. Then, we propose a Persona-Augmented Hypergraph Neural Network (PAHG) for PUID consequently. Extensive experiments are conducted on several state-of-the-art methods and our method.
  Keywords: intent detection; hypergraph learning; large-scale modelling.
  
  
• Role-playing gamification-based educator career promotion system  
  by Tubagus Mohammad Akhriza, Indah Dwi Mumpuni 
  Abstract: This article offers a solution to a long-standing problem in higher education institutions in Indonesia, where the number of associate professors and professors is still low. Lack of motivation is suspected to be one of the causes. HRD management often arbitrarily applies a reward and punishment system to increase motivation, where the assumption is that educators can be regulated through material rewards; however, this actually causes educators to feel less humanised. On the other hand, rewarding and punishing are gamification mechanics that should bring a fun game atmosphere, not a feeling of insecurity in educators when completing a task. This article proposes a gamification approach to Indonesian Educator Career Promotion System (ECPS) as a solution, by redefining gamification mechanics that channels the educator's interest in completing tasks. The implementation of the gamified ECPS prototype shows the diverse interests of educators, in contrast to the assumptions of the traditional system.
  Keywords: educator career; gamification; higher education; role-playing game.
  
  
• A clustering allocation and scheduling analysis approach for multiprocessor- dependent real time tasks  
  by Faten Mrabet, Walid Karamti, Adel Mahfoudhi 
  Abstract: The ultimate objective in this paper is to propose a new method for dependent tasks clustering by considering both the inter-tasks communication cost, the inter-clusters communication cost (inter-calculation units), the precedence impact, and the execution cost. The optimal Munkres assignment algorithm is used for an optimal total execution cost. Task deadlines and their imposed precedence obligations are taken into consideration to lead a fast and safe exact scheduling analysis of each partition separately, while giving pertinent feedback. Experimental results highlight the effectiveness of the proposed approach by comparing it with optimal ones. The outcome shows better results in the total execution cost and gives exact scheduling analysis results.
  Keywords: multiprocessor real-time systems; dependent tasks; clustering and allocation; heterogeneous multiprocessor architecture; partitioned scheduling; scheduling analysis.
  
  
• Simulation of photovoltaic systems artificial intelligence controller based on fuzzy perturbation algorithms  
  by Wei Liming, Li Kaikai, Wu Yangyun 
  Abstract: Solar energy is popular because it is clean and non-polluting. However, solar power generations efficiency restricts the development of the photovoltaic industry. To improve this problem, the research of maximum power point tracking (MPPT) algorithm is discussed by scholars. The paper proposes an artificial intelligence algorithm that is composed of fuzzy logic and perturbation observation method, so as to achieve the goal of MPPT. In the process of approaching the maximum power point, the optimal variable-steps size at each moment is adjusted by the fuzzy controller, which imitates the human brain. The simulation is carried out in Matlab/Simulink and is compared with the traditional perturbation observation algorithm. The result shows that the power loss of the artificial intelligence algorithm is significantly less than that of the traditional algorithm, and its strain ability is stronger when the environment changes abruptly, and the time to reach the maximum power point is effectively reduced.
  Keywords: photovoltaic system; MPPT; Fuzzy perturbation algorithm; perturbation observation.
  
  
• Intelligent machine vision model for building architectural style classification based on deep learning  
  by Aaron Rababaah, Alaa Rababah 
  Abstract: This paper presents an intelligent model for building architectural style classification. Image classification of architectural style is challenging to traditional machine vision methods. The main challenge in these systems is the feature extraction phase as there are many visual features in these styles that need to be extracted, refined and optimized. All these operations are done at the researcher discretion in traditional Machine Learning (ML) models. The advancements of ML to Deep Learning (DL) made automation of all the challenging operations possible. We constructed a machine vision model based on DL to investigate the effectiveness of DL in the classification problem at hand. A publicly available annotated dataset was used to train and validate the proposed model. The dataset consists of more than 5000 images of eight different architectural styles. The experimental results showed that the proposed model is reliable as it produced a classification accuracy of 95.44%.
  Keywords: architectural styles classification; machine intelligence; machine vision; deep learning; feature extraction; impact of number kernels/features.
  
  
• Assistive typing technologies: a new method based on binary sequence  
  by Luiz Augusto Costa, Geraldo Filho, Rodrigo Bonacin, Rodolfo Meneguette, Vinícius Gonçalves 
  Abstract: Users with severe motor difficulties may experience problems when interacting with traditional computing devices. This article addresses these problems by setting out a new method that involves interacting through binary sequences. A hardware and software system was designed based on this method. This system enables users to interact using binary movements that are triggered by clicking a single button. Comparative tests were carried out between the proposed method and Switch Control, an assistive technology embedded in iOS devices for the interaction of users with motor difficulties. In terms of typing speed, the results show that the proposed methodology proved to be more efficient in all the tested cases, and it was, on average, 22% more productive than the Switch Control.
  Keywords: communication; binary sequence; assistive technology; simulation; motor disabilities.
  
  
• Modelling and verification of enterprise application integration processes through coloured Petri nets  
  by Alexsandro Q. Lencina, Fabricia Roos-Frantz, Rafael Z. Frantz, Sandro Sawicki 
  Abstract: Integration processes are considered workflow processes that can be modelled using a process modelling language. Integration platforms usually offer a modelling language based on the well-known integration patterns. To automatically verify the logical correctness of such processes, they must be formally specified. Although there is a proposal in the literature that translate the integration patterns documented by Hohpe and Woolf into coloured Petri nets, it targets conceptual pattern definitions and does not cover all integration patterns found in actual implementation of integration platforms. More complex conceptual patterns usually have to be adapted when realised by a specific integration platform. We show how an actual modelling language that implements integration patterns can be translated into coloured Petri nets and its formal properties be verified. We intend to inspire researchers interested in the verification of models designed with other platform dependent message-based languages that also realise the same integration patterns.
  Keywords: coloured Petri nets; integration patterns; integration process modelling; model verification.
  
  
• A practical approach for the porting of the Ravenscar profile from ADA to C: method, rules adaptation and supporting tools  
  by Claudia Rinaldi, Fabio Romano, Paolo Serri, Luca Tiberi 
  Abstract: While conceiving with hard real time systems, determinism is the main requirement that must be satisfied in order for properly predict their behavior as required by their definition. For achieving this purpose an available solution is restricting Ada language tasking features to the Ada Ravenscar profile subset. This paper presents a solution to apply the Ravenscar profile concepts in systems where the tasking management is based on RTEMS real-time operating system, and C and Ada languages are used together. Moreover, a SW tool to automatically check the compliance with Ravenscar is proposed and the outcomes of some experimental activities proving the effectiveness of the SW tool are discussed
  Keywords: Ravenscar profile; Ada; C; RTEMS.
  
  
• Implementation of an IoT system for environment monitoring and remote web control using ARM Mbed cloud and GUI  
  by Shensheng Tang, Yi Zheng 
  Abstract: This paper implements an IoT system for environment monitoring and remote web control using ARM Mbed cloud. The embedded system used for environment monitoring is implemented using an ARM Cortex-M4 core-based STM32L4 series development board integrated with multiple sensors. The sensed data (i.e., temperature, relative humidity and atmospheric pressure) can be wirelessly sent to the Mbed cloud managed by the Pelion device management platform. The data values can also be sent to a graphical user interface (GUI) with user authentication. We develop three application appliances that can be controlled remotely over Internet through the Pelion platform. The proposed IoT system has been successfully implemented on the STM32L4 series development board with the main application program developed using C++ and the GUI developed by C# programming. The work of hardware and software co-design can be a practical paradigm of engineering education for IoT hobbyists and college students.
  Keywords: IoT; ARM Mbed cloud; Pelion Device Management Platform; GUI; C++; C#; WiFi; environment monitoring; web control.
  
  
• A method for evaluating the relationship strength of group users based on co-occurrence  
  by X.U. Jingke, XIAO Fei 
  Abstract: At present, the inference of users' social relations based on check-in data has become a hot topic. However, the existing methods to evaluate users relationship strength mainly concentrated between two users, and group users contain greater information and use value. This paper focuses on the co-occurrence phenomenon that group users are in the same time and space, and research group users relationship strength problem with co-occurrence method and min-max method based on entropy. The real check-in data sets are used to carry out sufficient experiments on the two proposed evaluation methods. The experimental results show that the min-max method can better adapt to the sparse problem of check-in data and reduce the influence of coincidence. Compared with the basic co-occurrence method, it has certain stability in groups of different sizes, and the larger the group size is, the more obvious its advantages are.
  Keywords: check-in data; group users; relationship strength; co-occurrence in time and space.
  
  
• Effects of big data analytics capability on performance of internet enterprises: chain mediating effects of strategic flexibility and strategic innovation  
  by Hua Zhang, Lifang Wang, Hongji Yang, Chunyuan Yu, Fubin Xia, Xinzhe Xue 
  Abstract: The recent interest in big data has led many companies to develop big data analytics capability (BDAC) in order to enhance firm performance (FP). However, BDAC pays off for some companies but not for others. It appears that very few have achieved a big impact through big data. To address this challenge, this study proposes a BDAC model drawing on the resource-based theory and the dynamic capability theory. In order to carry out the research, this paper takes Chinese Internet enterprises as the research object and obtains survey data from 629 employees through questionnaires. Through the test of the proposed chain mediation model using bootstrapped, it is found that: (1) big data analytics capability has significant positive influences on firm performance of Internet enterprises. (2) Strategic flexibility and strategic innovation play chain mediating roles on the path joining big data analytics capability and firm performance.
  Keywords: big data analytics capability; firm performance; internet enterprise; strategic flexibility; strategic innovation; chain mediating effects.
  
  
• Optimisation of energy consumption in cloud video surveillance centre based on monitoring and placement of virtual machines  
  by Majid Heidary, Ehsan Sadeghi Pour, Azad Noori, Maedeh Abedini Bagha 
  Abstract: Cloud computing is one of the most popular computational models, which requires plenty of physical devices where services are provided based on user demand; A majority of data centres need plenty of energy consumption which has become a challenge in recent years. Regarding cloud video surveillance as a contemporary research field of cloud computing and big data, the service, due to the high demand for monitoring remote places, continually consumes surplus energy to process the high-volume data. This study considers the importance of energy consumption in cloud video surveillance, and it has been tried to increase the efficiency of servers concerning energy usage. The proposed method employs virtual machine placement in two steps, including monitoring and placement, to reduce energy consumption and increase the efficiency of servers. Implementation results in Cloudsim showed that it reduces energy consumption and increases resource efficiency.
  Keywords: cloud video surveillance; virtual machine placement; energy.
  
  
• APP test system: a case study of calculus  
  by Ting-sheng Weng 
  Abstract: COVID-19 has had a broad impact on society, and a profound impact on education, thus, distance online courses are seen as a way to continue schooling during the pandemic. This study employed Android Studio to develop an APP calculus learning test system which can be used for self-review exercises and allows students to make good use of mobile apps to conduct post-learning and self-testing of calculus at home, and immediately determine their learning results. In addition, through back-end access, teachers can view students' learning scores and the number of wrong and correct questions, and thus, know the effect of individual students' self-review. During the COVID-19 pandemic, as teachers and students cannot interact in class at school, teachers can use the APP calculus learning test system to provide distance remedial teaching to students who fall behind during the course.
  Keywords: APP; test system; calculus; distance learning; COVID-19.
  
  
• Visual-inertial fusion positioning and mapping method based on point-line features  
  by Qinghua Feng 
  Abstract: In order to solve the problem of current visual SLAM (Simultaneous Localization And Mapping, SLAM) based on point feature technique in a structured or weak texture environment exist the problem of the strong dependence and weak noise immunity. This paper proposes a method of simultaneous location and map building based on the information of the visual point, line features and inertial information fusion. Firstly, feature points and feature lines are extracted by camera and fused with inertial information. Then recover the object's motion state in real time according to the information. The evaluations on the public dataset of EuRoc shows that the Root Mean Square Error (RMSE) is 0.152m and the method has excellent accuracy, robustness and timeliness.
  Keywords: SLAM; line feature; inertial information; recover; computer vision.
  
  
• An underwater vehicle odometry scheme based on visual-inertial fusion  
  by Yufan Wang 
  Abstract: In order to solve the problems of underwater vehicle inaccurate odometry caused by uneven illumination, current floating and blurred vision in underwater vehicle perception, this paper proposes an underwater vehicle odometry scheme based on visual-inertial fusion. The camera and IMU are first calibrated through the calibration board to solve the parameters of the device, and the accuracy of the pose can be ensured. Then, the visual-inertial odometry (VIO) system of underwater vehicle is designed to complete the fusion of camera information and IMU data. At the same time, an underwater shadow detection and removal algorithm is proposed to optimise the underwater information acquisition of the robot. Finally, this scheme is tested in various underwater data sets and different mainstream algorithms. The experimental results prove that the accuracy of the proposed method, and it has good robustness and pervasiveness in complex underwater scenes.
  Keywords: feature extraction; visual-inertial fusion; monocular camera; inertial measurement unit.
  
  
• Lightweight crop pest identification algorithm under natural background  
  by Dong Benzhi, Wang Yaqi, Xu Dali 
  Abstract: Aiming at the problem of poor detection effect and low recognition accuracy of small target insects under the background of complex natural environment, proposes an improved Yolo v5s insect recognition algorithm. The channel attention mechanism is embedded in the backbone network. The adaptive spatial feature fusion (ASFF) structure is introduced in the PANet part, and dynamic weight parameters is used to assign different weights to feature maps of different scales, Finally, we change the loss function and non-maximum suppression strategy to improve the accuracy of bounding box positioning and the speed of regression. Experimental results show that the improved algorithm has a final average accuracy (mAP@0.5) of 97.8% in the D0 dataset and an average detection speed of 13.66 ms per image, which is more suitable for deployment in mobile and embedded devices to achieve real-time detection.
  Keywords: insect recognition; lightweight convolutional neural network; channel attention mechanism; adaptive spatial feature fusion.
  
  
• Enhancing the accuracy of transformer-based embeddings for sentiment analysis in social big data  
  by Wiem Zemzem, Moncef Tagina 
  Abstract: Social media have opened a venue for online users to post and share their opinions in different life aspects, which leads to big data. As a result, sentiment analysis has become a fast-growing field of research in natural language processing (NLP) owing to its central role in analysing the public's opinion in many areas, including advertising, business, and marketing. This study proposes a transformer-based approach, which integrates contextualized words with Part-Of-Speech (POS) embedding. Then, the enhanced word vector is forwarded to a hybrid deep learning architecture combining a Convolutional Neural Network (CNN) and a Bidirectional Long Short Term Memory (BiLSTM) to discover the post's sentiment. Extensive experiments on four review datasets from diverse domains demonstrate that the proposed method outperforms other machine learning approaches in terms of accuracy.
  Keywords: deep learning; sentiment analysis; word embedding; big data; natural language processing.
  
  
• Methodology of aircraft structural design optimisation  
  by Nihong Yang 
  Abstract: Aircraft structural design normally aims to achieve the lightest weight while meeting aircraft performance requirements. Various optimisation methods and approaches are used in structural design optimisation. This paper reviews the aircraft structural optimisation process and methods currently applied in aircraft structural design, including structural analysis and optimisation process, FEA modelling techniques and optimisation algorithms. It is seen that a two-step global local approach is extensively applied in aircraft structural optimisation. Submodelling technique is often used in the global local optimisation, whereas superelement technique can provide accurate solutions. Gradient based optimisation algorithms are suitable for optimisation problems with continuous variables such as size and distance of metallic aircraft structures, whilst genetic algorithms and particle swarm optimisation algorithms are often used for optimisation problems with discrete variables such as laminate composite thickness and stacking sequence.
  Keywords: aircraft structures; design optimisation; FEA modelling techniques; laminated composites; numerical modelling; optimisation algorithms.
  
  
• Unsupervised machine learning schemes for cooperative spectrum sensing in cognitive radio  
  by RAJENDRA YELALWAR, YERRAM RAVINDER 
  Abstract: The major challenge in the development of recent wireless technology is spectrum scarcity which is addressed by introducing the Cognitive Radio (CR) technique. In CR, spectrum sensing is the most critical task that senses the surrounding environment to detect the presence of a primary User (PU) in the target spectrum. This paper proposes the machine learning (ML) enabled Cooperative Spectrum Sensing (CSS) approaches where the application of clustering algorithms for the eigenvalue based CSS under different fading channel conditions is explored. The sensing performance is analysed with different PUs, signal features, Signal to Noise Ratio (SNR) values, and channel conditions. Secondly, this work proposes the novel clustering based CSS framework for Non-orthogonal Multiple Access (NOMA) signal detection. The system performance is measured in terms of sensing accuracy and Receiver Operating Curve (ROC). The simulation results ensure the effectiveness of the proposed clustering based CSS framework compared to the existing work in terms of improved accuracy which is observed to be 92.5% for K means clustering based CSS framework for NOMA
  Keywords: spectrum sensing; machine learning; K-means; K-medoids; agglomerative; NOMA.
  
  
• FPGA implementation and Multisim simulation of a new four-dimensional two-scroll hyperchaotic system with coexisting attractors  
  by Sundarapandian Vaidyanathan, Esteban Tlelo-Cuautle, Khaled Benkouider, Aceng Sambas, Ciro Fabian Bermudez-Marquez, Samy Abdelwahab Safaan 
  Abstract: Field-programmable gate array (FPGA) design of a new four-dimensional two-scroll hyperchaotic system is investigated in this work. A detailed system modelling of the new system with a hyperchaotic attractor begins this work with phase plots, which is followed by a bifurcation study of the new system. Special dynamic properties such as multistability and symmetry are also investigated for the new system. Using Multisim software, a circuit model is designed and simulated for the new hyperchaotic system. FPGA design and Multisim simulation of the new system enable practical applications in science and engineering. The implementation of the FPGA design in this work is carried out by applying two numerical schemes, viz. Forward Euler and Trapezoidal methods. Experimental attractors observed in the oscilloscope show good match with the Matlab signal plots.The FPGA hardware resources are detailed for both numerical methods.
  Keywords: hyperchaos; bifurcation; symmetry; phase plots; hyperchaotic system;rnparameters; stability; multistability; circuit model; FPGA implementation.
  
  
• A study of integration application based on 5G/8K/AI/VR for the activation of intangible cultural heritage  
  by Lu Zhang, Shaojun Ji, Meiyu Shi 
  Abstract: Within the culture and tourism field, the activation of intangible cultural heritage has been a hot topic in recent years. A major way of doing that is through leveraging various cutting-edge technologies, in particular, 5G/8K/AI/VR. However, due to the diversity of intangible culture and inescapable relationships among different technologies, it is necessary to consider the integration application of the advanced technologies. There is a trend that the said technologies are often used to construct the immersive experience. A related embodied cognition theory is referred in this paper to build the theoretical basis. Then, a detailed analysis of the functions and features of 5G/8K/AI/VR is followed. Based on the research efforts, several integration application scenarios are summarized through case study for the activation of intangible cultural heritage.
  Keywords: integration application; 5G/8K/AI/VR; activation; intangible cultural heritage.
  
  
• A systematic mapping study on IoT-based software systems for precision agriculture  
  by Vinícius Lopes, Cleiton Silva, Dayana Gonçalves, Roberto Oliveira, Renato Bulcão-Neto, Mohamad Kassab, Valdemar Graciano-Neto 
  Abstract: Context: Agriculture is often pressured to adopt new technologies so that production rate can be accordingly increased. The Internet of Things (IoT) has played an essential role in modernizing the agricultural practices once it can support monitoring and automated decisions on planting. Objective: Given the ascending adoption of IoT in agriculture, the main goal of this study is reporting collected evidence from the literature and summarize how IoT systems have been used to support several activities in the agriculture domain. Method: We adopted the systematic mapping study (SMS) procedure. We designed a search string executed in two Web search engines: Scopus and Embrapa repositories. We selected a total of 35 primary studies that either propose or evaluate IoT-based systems in the agriculture domain. Results: Results suggest that, although there are different platform solutions such as Web-based systems, Web-and-mobile-based systems, and mobile systems targeting activities such as planting monitoring and irrigation, there are still several opportunities in the area, such as (i) the conception of automated decision-making processes and supporting technologies for agriculture recurrent activities, (ii) adoption of edge computing and machine learning for information processing and automation, respectively, and (iii) proposition of solutions towards reducing production costs and ecological impacts.
  Keywords: software system; internet of things; precision agriculture; 4.0 agriculture, smart agriculture; mapping study.
  
  
• Experimental study for makespan reduction in enterprise application integration processes using bio-inspired algorithms  
  by Maira S. Brigo, Fernando Parahyba, Rafael Z. Frantz, Sandro Sawicki, Fabricia Roos-Frantz 
  Abstract: Enterprise Application Integration area seeks to support the companies' business processes by enabling data and functionality of the applications to become reusable. Integration platforms are tools that develop and execute integration processes. This execution is done by a key component of the platforms called run-time system; that said, the performance from integration processes heavily depends on the efficiency of the run-time system. The task-based execution model implemented by the run-time system can use a strategy based on local pools to store computational threads associated with each task that make up the workflow of the integration process, to execute them. The challenge in this strategy is to evenly distribute the threads in each pool, minimising the makespan. We propose a experimental study, which uses two meta-heuristics to find the best distribution with the optimal number of threads. We compared both Particle Swarm Optimisation and Cat Swarm Optimisation, with the latter showing better results.
  Keywords: makespan; task-based; run-time system; optimisation; integration platforms; integration process; meta-heuristics; particle swarm optimisation; cat swarm optimisation; threads.
  
  
• Risk assessment of construction safety of prefabricated building hoisting based on cloud model-entropy method  
  by Chengkuan Fang, Chunling Zhong, Yunlong Zhang 
  Abstract: In order to ensure the safety of assembly building hoisting construction, the safety risk assessment index and model of assembly construction are studied. Based on assembly building hoisting construction characteristics, accident inducement, field investigation, and literature research, the safety risk assessment model of assembly building hoisting construction based on the cloud model and entropy weight method is constructed. Using the cloud model effectively reduces the fuzziness and randomness of risk assessment data, and then using the entropy weight method, the weight coefficient of the evaluation index is given, which effectively avoids the subjectivity of expert weighting. The model is applied to engineering examples, and good results are achieved. The results show that the risk level evaluation results of the model are consistent with the field risk level, indicating that the model can objectively and accurately evaluate the risk level of assembly building hoisting construction.
  Keywords: prefabricated building; hoisting; cloud model; entropy method; risk assessment.
  
  
• DL-RED: a RED-based algorithm for routers  
  by Samuel Hassan, A. Rufai, C. Ajaegbu, F. Ayankoya 
  Abstract: Keeping the average queue size small (which will in turn, offers a minimised delay performance) is regarded as an important goal of Active Queue Management (AQM) algorithms implemented in internet routers. The Random Early Detection (RED) algorithm is unable to achieve this desired goal. In this paper, we present an enhanced RED algorithm, named Double Linear RED (DL-RED) which uses a linear packet dropping function for a light - and moderate - network traffic load scenarios, and another linear packet dropping function for a heavy network traffic load scenario. The effectiveness of DL-RED was evaluated and compared with RED and Nonlinear RED (NLRED) algorithms using ns-3 simulation tool. Experimental results proved that DL-RED clearly performed better than RED and NLRED with reference to delay and throughput. A little effort is required to amend the packet dropping probability profile of RED's algorithm implementation with DL-RED algorithm. Therefore, RED can be easily replaced/upgraded in internet routers with DL-RED.
  Keywords: AQM; congestion control; delay; DL-RED; simulation.
  DOI: 10.1504/IJCAT.2022.10053495
   
  
• Improving recentness of the ICT book recommendation using an adaptive rules-based recommender system  
  by Mochammad Husni, Tubagus Mohammad Akhriza, Sarifuddin Madenda, Eri Prasetyo Wibowo 
  Abstract: The traditional library book recommendation system (RS) has limitations where all recommendations only refer to internal book borrowing transactions; while the development of science and technology, especially in the field of ICT, has exceeded the theme of the recommended books. As a result, the recentness of the recommendations is questionable. As a solution, this article proposes an adaptive-rules-based book RS while at the same time introducing a dimension to measure the quality of recommendations namely recentness. It measures how up-to-date the recommended book theme is, against a set of trending themes extracted from external publications. An experiment was conducted to measure the book recommendations generated by the new RS in a library, compared to a collection of recent publications in the IEEE Xplore database. At first, the recentness of the recommendation was only around 23.5-57.1%, but it was successfully increased to 47.6-76.2% by the proposed RS.
  Keywords: association rule; library; recommendation quality; recommendation system.
  
  
• Cooperative game amongst prefabricated building chain stakeholders based on improved Shapley value method  
  by Qi Zhao, Chunling Zhong 
  Abstract: The cooperation among stakeholders at all nodes is the premise for the stable operation of the prefabricated building chain. Such cooperation is the key factor to the stable development of the prefabricated building industry. The cooperation among stakeholders cannot be separated from a reasonable and effective benefit distribution mechanism. Firstly, the prefabricated building chain is defined to analyse the cooperative relationship among stakeholders. Secondly, the study investigates the influencing factors of interest distribution among all stakeholders of the prefabricating project and puts forward the distribution principles. Finally, profit distribution using the Shapley value method of the prefabricated building chain model is performed on the basis of cooperative game theory. At the same time, it considers the influencing factors of prefabricated industrial chain profit distribution using the entropy weight method to improve the model. Particularly, it improves the rationality and objectivity of the profit distribution model and makes up for the theoretical vacancy of benefit distribution in the prefabricated construction industry chain.
  Keywords: prefabricated building industry chain; stakeholder; Shapley value method; entropy weight method; cooperative game.
  
  
• A comprehensive review of clustering approaches for energy efficiency in wireless sensor networks  
  by Wesal Bassem Nedham, Ali Kadhum M. Al-Qurabat 
  Abstract: Wireless Sensor Networks (WSNs) have become more popular in recent years due to their vast range of applications. The use of WSNs is an absolute requirement for future revolutionary domains such as smart cities, the Internet of Things, or ecological fields, where hundreds or thousands of sensor nodes are placed. Moreover, because WSNs are energy-constrained networks, implementing energy-aware protocols is critical. Hierarchical techniques enhance network performance and extend network lifetime in large-scale WSNs. Within a WSN, hierarchy is achieved by dividing the network into sub-networks known as clusters, which are directed by Cluster Heads (CH). Clustering is the most common energy-efficient approach, and it offers several benefits, such as reduced latency, scalability, lifetime, and energy efficiency. This study presents a detailed assessment of several clustering techniques, together with their aims, features, etc. Furthermore, clustering techniques are classified and evaluated based on numerous cluster features, cluster head attributes, and clustering procedures.
  Keywords: wireless sensor networks; energy consumption; energy efficiency; clustering techniques; IoT.
  
  
• A novel shape-based time series classification with SAX-ensemble  
  by Mariem Taktak, Slim Triki 
  Abstract: Since the first publication of the Symbolic-Aggregate approXimation (SAX), many extensions with novel SAX-distance measure have been published. Each of them attempts to integrate additional statistical features in order to improve original SAX average-based feature. Each SAX-feature has its own distance function which quantify the (dis)similarity between two Time Series (TS). However, none of them can fit the overall shape-characteristics of a TS and give the superiority to an individual SAX-based classifier. In order to combine the prediction of each single SAX-based classifier, we propose a collection of several SAX-feature to compose a shape-based ensemble for TS classification. The proposed SAX-Ensemble scheme is applied on a multiple domain representation of the TS where the diversity of collected SAX-feature make the setting of the SAX-discretization parameters a challenging task especially for a long TS data or a large training dataset. In order to avoid a time-consuming of either grid search or expensive optimization algorithm, we instead apply a data-aware or data-agnostic parameters setting technique. Experimental results on real TS database show that the performance of the proposed SAX-Ensemble with data-aware technique exceeded the SAX-based classifiers with more flexible and realistic parameters estimation.
  Keywords: time series data; symbolic aggregate approximation; SAX; shape-based classification.
  
  
• A concrete product derivation in software product line engineering: a practical approach  
  by Karam Ignaim, Khalid Alkharabsheh, André L. Ferreira, João M. Fernandes 
  Abstract: Software Product Lines (SPLs) support the development of a full family of products through systematic reuse of shared assets. Product derivation is a key activity in SPL engineering and one of the primary issues that an SPL faces. This paper presents a practical approach that supports an automated derivation of a product from an SPL. The SPL is derived from a family of products that originated from a non structured approach to manage variability. The automated derivation approach is based on the use of product configurations and Feature Models (FMs) refactoring. The approach was deployed and evaluated with a real-world SPL in the automotive domain. The result reveals that the approach derives a product in an automated and successful way.
  Keywords: software product line; product derivation; feature models; product configuration; refactoring.
  DOI: 10.1504/IJCAT.2022.10054830
   
  
• A Comprehensive Review of the Electroencephalography Data Analytics  
  by Marwa Saieed Khlief, Ali Kadhum Idrees 
  Abstract: This paper proposes a comprehensive review of Electroencephalography (EEG) data analytics. The EEG signal definition and the analysis process are presented. The public EEG datasets that were used by the researchers are explored. EEG data acquisition methods are investigated. This paper covers and summarises the work and techniques that have been done to compress EEG data. Significant approaches for feature extraction for EEG signal processing are illustrated. The collected features are then used to classify signals based on their properties. Machine learning techniques have become very important in this field in recent years because of their incredible ability to assess complicated volumes of data. Therefore, machine learning and deep learning for EEG data have been introduced. For researchers interested in EEG data analysis, this work can serve as a basic strategy and a roadmap.
  Keywords: electroencephalography; EEG signal processing; data compression; machine learning; deep learning.
  
  
• Performance analysis of machine learning algorithms applied to network intrusion detection  
  by Minyar Sassi Hidri, Suleiman Ali Alsaif, Adel Hidri 
  Abstract: Despite enormous efforts by researchers, Intrusion Detection Systems (IDSs) still face challenges in improving detection accuracy while reducing false alarm rates and in detecting novel intrusions. Recently, machine learning-based IDS systems are being deployed as potential solutions to detect intrusions across the network in an efficient manner. Most of them cannot perform well with large-scale or even real-time data, while the rest cannot track down evolving malicious attacks, thus putting a huge void in existing solutions. The proposed approach is an attempt to explore the possibility of developing an IDS which analyses raw network data in the form of network traffic files or server logs, allowing us to simulate a real environment to accomplish testing and evaluation. Thanks to several conducted experiments, we were able to demonstrate that it is possible to improve the overall performance of learning algorithms in the field of network security by model biasing.
  Keywords: machine learning; intrusion detection system; malicious attacks; model biasing; network traffic.
  
  
• MODEL AND SIMULATION OF TRACKING MOTION FOR ACCURATE WELDING USING A UNIVERSAL COBOT  
  by Ha Quang Thinh Ngo, Thanh Trung Thai, Hao An Nguyen, Vo Nhut Quang Mo, Quang Long Le, Tuan Anh Vu, Aly Mach 
  Abstract: Welding is a process that requires the skill and time of a professionally-trained welder. However, this welding performance is not highly productive. Therefore, in this paper, the control approach of a 6-DOF collaborative robot (Cobot) arm is investigated in the application of a MIG (metal inert gas). By studying the trajectory generation algorithm, the Cobot can be used to track the welding curve. A theoretical model of the 6-DOF Cobot arm is successfully established in 3D space. Then, the results to validate the method according to the welding trajectory are presented in numerical simulations. It can be clearly seen that our approach shows great position-controlled ability. From these achievements, the applications of this Cobot are expected to be implemented in many industrial fields.
  Keywords: robot arm; cobot; automatic welding; position control; motion tracking.
  
  
• Improving the routing performance in optical networks by an optimised BFS algorithm  
  by Juliano Antonio Da Silva, Djeisson Hoffmann Thomas, Fernando Cesar De Castro 
  Abstract: The procedure of routing and wavelength assignment became a limiting factor for new applications in optical networks (ONs) due to the faster dynamic and greater unpredictability of service requests. The need for quick service recovery in the event of failures is also a crucial aspect which depends on this procedure. The present work proposes a new assignment of alternative routes able to comply to any data service request in ONs. The assignment procedure is based on the distinction by low correlation between paths, on the available bandwidth of different paths and on the latency for the establishment of communication between network nodes. In such context, an improvement of the Breadth-First Search algorithm (BFS) is proposed and tested with parameters from a real world, resulting in a good performance to find alternative routes and to classify them according to the best score obtainable, independently of the source or destiny nodes.
  Keywords: optical networks; optical network performance optimisation; routing in optical networks; RWA; routing algorithms.
  
  
• Prescribed-time leader-following group consensus for linear multi-agent systems with delays  
  by Chen Xin, Dai Lingfei, Guo Liuxiao, Yang Yongqing 
  Abstract: This paper challenges the problem of prescribed time group consensus for linear multi-agent systems with delays. By designing the delayed protocols based on prescribed time scaling function, the multi-agent systems can realise group consensus in any preset convergence time, which is independent of both the initial conditions and system parameters. Most existing results on finite or fixed time bipartite consensus, require the considered structurally balanced of strongly connected signed networks. In this paper, prescribed time group consensus, including bipartite consensus as its special item, can be achieved without the assumption of signed balanced networks. In addition, group consensus according to different delays are also presented by resorting the Lyapunov stability and algebraic graph theory. Simulations illustrate the validity and correction of the proposed protocols.
  Keywords: prescribed-time; time delay; group consensus; multi-agent systems.
  
  
• Research of concept, model, construction technology and effectiveness of ubiquitous smart learning space  
  by Lei Song, Sirirat Petsangsri, Jirarat Sitthiworachart 
  Abstract: This paper aims to construct ubiquitous smart learning space to enhance learning skills. To ensure implementation effectiveness, this paper proposes the concept of ubiquitous smart learning space, gives a seven-layer architecture model, the functional orientation, construction principles and technical implementation strategies, through those strategies, the space model creatively has been realised and maximally meets learning stakeholders needs. To test its effectiveness, we surveyed 263 students and five teachers on satisfaction and self-efficacy assessment. The results show that it supports diversity and adaptability smart learning environment and facilitates learners to learn easily, engagingly, and effectively, and it also proves that collaborative learning activities guided by the USLS collaborative learning model significantly contribute to student learning and skills development. Teachers are more satisfied with collaborative learning activities guided by the model in ubiquitous smart learning space than the traditional classrooms.
  Keywords: learning space; smart learning; ubiquitous smart learning space concept; architectural model; learning effectiveness; construction technology.
  
  
• Automatic design of conformal cooling channels with an asymmetric centre  
  by Yuan-Ping Luh, Chien-Chuan Chin, Hong-Wai Iao 
  Abstract: In this paper, a compensation-based calculation method is proposed for designing a conformal cooling channel (CCC) with an asymmetric centre to make the temperatures of the core and cavity moulds uniform. By redefining the parting line, a heat accumulation centre can be obtained, and the distance between a cooling channel and a plastic part can be calculated to ensure a more uniform temperature distribution, thereby reducing the severity of volumetric shrinkage and deformation. Two case studies with different characteristics were used for verifying the proposed method. One is a flat parts had large thickness difference; the other is a uniform thickness with a large curved surface. According to the analysis results obtained in the aforementioned cases, the CCC with an asymmetric centre not only reduce the temperature of molten plastic rapidly to the ejection temperature, but also effectively evened the temperature variation over the entire mould cavity.
  Keywords: asymmetric centre; conformal cooling channel; automatic construction.
  
  
• Hybrid learning model for analysing the Uppal Earth region, in Telangana State using multispectral Landsat-8 OLI images  
  by P. Aruna Sri, Santhi Vaithiyanathan 
  Abstract: Remote Sensing (RS) and Geographical Information systems (GIS) are being widely used to carry out analysis of the earths surface with respect to changes in space and time. Remote Sensing and GIS permit us to extract useful information from the earths surface to understand land use and land coverage, monitoring urban growth, and detect different urban patterns. In this paper, a hybrid learning model is proposed for the classification and analysis of the Uppal earth region which is located nearby Hyderabad in Telangana state. In the hybrid learning model, the ISODATA clustering algorithm is combined with the Normalized Vegetation Index (NDVI) and K-means learning model. In this proposal, the spectral features of the Uppal region from the satellite images are used to carry out the analysis for the same region. The proposed hybrid learning model is tested with multispectral Landsat-8 OLI images of the Uppal earth region and the accuracy of the algorithm is measured. The accuracy of the proposed Merged-ISODATA algorithm gives 74.33 and the kappa value is 0.64. The obtained results of existing methods such as ISODATA clustering and K-Means algorithms are 71.5 and kappa value is 0.5 These values represent that obtained results of the proposed algorithm are better than the results obtained in existing approaches. It is elaborately presented in the results and discussion section
  Keywords: Landsat-8 OLI; remote sensing; normalised vegetation index; accuracy.
  
  
• Seeking a sustainable urbanisation path: the characters of interlink of Jilin Province within Ha-Chang Urban Agglomeration based on using rail and road networks big data analysis  
  by Bingxin Li, Shuying Cao, Tong Liu, Lei Liu 
  Abstract: Since Jilin Province serves as the most convenient link between Liaoning and Heilongjiang Province, its enhancement of regional spatial linkage holds the prospects of Haerbin-Changchun urban agglomeration in China. The research aims to seek a sustainable urbanization path by analysing the characters of interlink of Jilin Province with the help of GIS based space syntax methodology, by using transit network data with higher resolution, broader coverage and finer representative. The results show that (1) the spatial accessibility among Dunhua, Fusong and Hunchun city should be prioritized to narrow down the spatial inequality among leading cities. (2) the regional connectivity of Shuangliao-Jian belt should be given more attention compared to other belts. (3) as the topological forms and national boundaries have direct influence upon city clusters, the southern cluster strengthen its outward connection with Liaoning Province, while the eastern one should include Antu-Yanji, Tumen and Antu city to enhance its inward connection with cities along north-south direction.
  Keywords: big data; regional spatial linkage; GIS-based space syntax method; urbanization development.
  
  
• Mathematical modelling pricing of instances the providers of the infrastructure as a service: an adapted hedonic approach  
  by Laize D. L. Trindade, Fernando Parahyba, Rafael Z. Frantz, Fabricia C. Roos-Frantz, Sandro Sawicki 
  Abstract: In Infrastructure-as-a-Service (IaaS), the large number of providers and instances combined with the lack of transparency in understanding pricing policies hampers the choice for the best provider-plan-instance combination. An issue less explored by the scientific community is the physical location of data centre and its influence on forming the final price of an instance. We developed a set of four mathematical models that estimates and represents in detail the prices of each of the three leading provider in the market and the canonical model. The modelling was empirically carried out. The verification and analysis of the individual models, and of the canonical model, were carried out through the comparison of the real price and the price estimated by the model. Thanks to that, it was possible to see that the values obtained represented characteristics from computational resources in the composition of the provider plans.
  Keywords: cloud computing; infrastructure-as-a-service; canonical model; mathematical modelling; hedonic model; IaaS providers; geographic location; pricing; estimate.
  
  
• A review of current prediction techniques for extending the lifetime of wireless sensor networks  
  by Wesal Bassem Nedham, Ali Kadhum M. Al-Qurabat 
  Abstract: The possibility for broad usage of wireless sensor networks (WSNs) in many various sectors, such as environmental monitoring, security, home automation, and many others, has increased research interest in WSNs. Although its successes, the broad proliferation of WSNs, especially in distant and inhospitable areas where their usage is most advantageous, is hindered by the primary obstacle of limited energy, as they are often battery operated. To provide these energy-hungry sensor nodes with a longer life expectancy, one technique to achieve this aim is to reduce the frequency of data transfer. Conversely, a portion of the observed data could be predicted to avoid initiating communications that might overwhelm the wireless channel. In this paper, we classify and analyse current prediction-based data reduction strategies for WSNs. Our key contribution is a systematic technique for choosing a prediction model in WSNs based on WSN limitations, prediction technique features, and observed data.
  Keywords: wireless sensor networks; prediction models; time series models.
  
  
• Lightweight design and realisation of autonomously balancing bicycles based on additive manufacturing technology  
  by Yizhi Wang, Yimeng Zhang, Bing Wang, Zhong Yang, Xingliu Hu, Lingyi Huang, Yang Zhang 
  Abstract: Lightweight design based on additive manufacturing (AM) technology enlightens the design and control of autonomously balancing bicycles. However, apart from increased energy efficiency by weight reduction, static and dynamic stability of bicycle body is degraded, which becomes a significant challenge of control system design. Therefore, in this research, taking autonomously balancing bicycle as a prototype, firstly alternative parts made in ABS, aluminium alloy and nylon through AM technology are proposed and realised. Experiment results show, compared to original design in stainless steel, 36.3% weight is reduced, mechanical characteristics are acceptable but stability is significantly degraded with the ABS-based alternatives. Further, the data fusion method was creatively introduced to control system design to pre-processing the data from gyroscope and accelerator. After tuning process, the designed method obtained a smoother control performance with faster dynamics to compensate the degradation of static and dynamic stability.
  Keywords: lightweight design; additive manufacturing; autonomously balance; control system design; data fusion method.
  
  
• Mobile application for the management and control of construction site safety: a comparative study  
  by Yahel Giat, Amichai Mitelman 
  Abstract: We investigate whether a mobile application for the management and control of construction sites induces the management of safety. Four construction sites with the application integrated into their operation are compared with four similar sites without the application. A five-day survey was conducted in each site to measure various safety related features and to interview and observe site managers. The results show that except for two variables, the sites with the application exhibit safer behaviours implying that the application improves the management of safety in these sites. We find that users view the application as a safety management tool and not just as a record keeper or as a means for legal protection. However, they will avoid using important built-in safety features if they feel that these take too much time. In these cases, they will continue to use traditional tools despite their limited effectiveness.
  Keywords: mobile devices; global firms; construction; safety management; information management; mobile application; control; IT.
  
  
• Electronic circuit design and FPGA-based hardware implementation of a new multistable 4-D hyperchaotic four-leaf system  
  by Sundarapandian Vaidyanathan, Aceng Sambas, Daniel Clemente-Lopez, Jesus M. Munoz-Pacheco, Alain Soup Tewa Kammogne, Vannick Fopa Mawamba, Siewe Siewe Martin, Samy Abdelwahab Safaan 
  Abstract: We generate a new hyperchaotic four-leaf system by inserting a state feedback control to the 3-D Vaidyanathan-Rhif chaotic four-leaf system (Vaidyanathan and Rhif, 2017). The new hyperchaotic system developed in this work has three quadratic nonlinearities and two positive Lyapunov exponents. Bifurcation analysis points coexisting bifurcations and multiple attractors for the new 4-D hyperchaotic four{-}leaf system. Next, we construct an electronic circuit model using MultiSim (Version 13.0) for implementing the new 4-D hyperchaotic four-leaf system. Embedded and non-embedded implementations of chaotic and hyperchaotic systems are vital to increase their usability in many engineering applications since those FPGA designs can be linked to the digital world surrounding us in a direct manner. An ARM-FPGA-based hardware implementation of the 4-D hyperchaotic four-leaf system is given in detail, showing a good agreement between numerical and experimental results.
  Keywords: hyperchaos; chaos; electronics; circuit design; stability; FPGA design; hyperchaotic systems; phase portraits; bifurcation; system analysis.
  
  
• Modelling and performance analysis of a cloud computing system using an open queueing network with multi-server queues  
  by Shensheng Tang 
  Abstract: A queueing model of an open queueing network with multi-server queues is developed for a cloud computing system and the performance is analysed based on different system parameters such as the arrival and service rates, the number of servers in each node, the routing probability, and the number of service types. By using the Laplace-Stieltjes transform (LST) technique, the close-form expression of the cumulative distribution function (CDF) of the total response time is derived. On this basis, some interesting application scenarios are discussed. Detailed numerical evaluations of the developed performance metrics are conducted and simulation results are provided for the application scenario verification. The proposed system modelling and performance analysis method is expected to provide a useful reference for the design and evaluation of cloud computing systems.
  Keywords: cloud computing; web server; service centre; queueing network; multi-server queue; Laplace-Stieltjes transform; cumulative distribution function.
  
  
• A secure and effective diffused framework for intelligent routing in transportation systems  
  by N. Bharathiraja, M. Shobana, M. Vijay Anand, R. Lathamanju, C. Shanmuganathan, V. Arulkumar 
  Abstract: The consistently expanding traffic, different postponement delicate administrations, and energy use compelled prerequisites have carried gigantic difficulties to the ongoing correspondence networks in the transportation framework. Because of the great speed and repeating topological variations of vehicular sensor networks, determining an associated course with sufficient idleness is a difficult task with many requirements and barriers. As a result, in order to combat this, we developed a measurable method for dealing with presumably determining the heap clog and energy use during the lifespan of the sensor network for transportation framework. The paper proposes a secure and effective diffused framework that highlights lower energy use and secure correspondence. The least bounce included in coordinated dissemination is used as the rule for developing the slope in this system, which further develops security and dependability by adding the possibility of the angle to flag the course and pace of information transmission.
  Keywords: secure; reliability; routing algorithm; transport systems; intelligent routing; sensor networks.
  
  
• A novel approach to secure biometric data using integer wavelet transform, chaotic sequences and improved logistic system-based watermarking  
  by Payal Garg, Ajit Jain 
  Abstract: Watermarking can be used for security as no one carries files or documents around anymore. To retain integrity a watermarking algorithm is applied that uses integer wavelet transform (IWT), chaotic sequences for biometric images in an improved logistic system (ILS) which must be of high capacity and Least Significant Bits (ICIL). In such schemes, users can see some distortion, which is objectionable. In standard, insertion of the watermark is done by changing the grey level of some of the pixels but is not effective against robust attacks. By this algorithm, one can achieve high-level security using chaotic sequences for watermark encryption. The proposed work was tested using datasets such as CASIA-v5 and Kaggle-Fundu and evaluated on various parameters, namely, PSNR, NC, and BER. When compared to others, a high imperceptibility with an average PSNR of 52.39 dB is significantly more resilient against different attacks.
  Keywords: watermarking; biometrics; medical data security; data embedding; chaotic maps; IWS.
  
  
• Recent trends in intelligent transportation systems using big data analysis  
  by Mahvash Iftikhar, Zain Anwar Ali, Muhammad Shafiq, Arsalan Lodhi 
  Abstract: The fusion of technologies introduces the various applications of smart cities. Internet of Things (IoT) is the core concept for smart cities. The foremost functionality of IoT is to access the data collected from the shared devices over the communication technologies from the sensors. The data from the various sources are mostly heterogeneous in nature. With the Big data technology, the gaining more insight of the data is becoming possible. Big data organises the data, brings efficiency by increasing the performance. Big data works on the three Vs, that is enormous generation of data refer to as volume, speediness of data generation refer to as velocity and the generation of data in either structure or unstructured format that is variety. To take decision on the data is the evolving area of research. The data is then pre-processes with the big data technology and formerly use intelligent algorithms to take the real-time decision.
  Keywords: smart city; intelligent transportation system; AI algorithm; bio-inspired algorithm; computer vision; big data analytics.
  
  
• Research on visualisation algorithm of handwritten digital image recognition based on deep neural network  
  by Fang Teng, Xing Liu Hu 
  Abstract: The efficiency and accuracy of manual observations of Modified National Institute of Standards and Technology (MNIST) handwritten digital images are low. To solve this problem, a method based on a deep neural network (DNN) model is proposed for screening, classifying, and recognising handwritten digital images. MNIST handwritten digital images are used to train and test the DNN model for the rapid and accurate recognition. The average recognition accuracy of DNN model is 96.46%. The interactive interface is designed to realise the visualisation of programs and algorithms, and algorithms can be analysed from different angles and levels. From comparing the recognition effect of the DNN with local binary pattern (LBP) feature extraction using texture features and edge feature extraction using shape features, the experimental results show that the DNN not only has high recognition accuracy, but also simplifies the complex process for manually extracting image features.
  Keywords: visualisation; handwritten digital image; image recognition; deep neural network; local binary pattern feature extraction; edge feature extraction; numbers; image classification; machine learning; visual interface.
  
  
• An abstraction-based approach to eliciting wisdom in intangible cultural heritage utilisation  
  by Yu Wang, Sicong Ma, Lin Zou, Hongji Yang 
  Abstract: Intangible cultural heritage utilisation action involves a variety of knowledge, such as culture, history, politics and suchlike. However, intangible cultural heritage is arduous to meet the customers requirement in current situation. A challenge for intangible culture heritage barriers is to provide 'wise' products to satisfy the customers requirements. In order to achieve this target, this paper presents a novel system based on abstraction techniques, aiming to search or generate wisdom in the intangible cultural heritage domain. It covers three phases, i.e., Abstracting Data/Information/Knowledge, Searching Metadata and Ranking the wisdom results. Abstraction techniques aim to analyse the characteristics of customers requirements and construct the levels of abstraction. Cuckoo Search is used for searching relevant metadata and potential relevant meta data to express the wisdoms. Furthermore, wisdom metrics are built to rank the wise products.
  Keywords: intangible cultural heritage; abstraction techniques and wisdom; creative computing.
  
  
• Large scale orthogonal integer wavelet transform features based active support vector machine for multi-class face recognition  
  by Tanvi Dalal, Jyotsna Yadav 
  Abstract: Support vector machines are widely used in the field of face recognition (FR) but they suffer from the drawback of high computational time. In proposed work, new active set strategy is used for support vector machines on integer wavelet transform (IWT) based large scale facial features with low computational time. Lifting scheme based significant localised wavelet features are extracted using IWT based on orthogonal wavelets. Large Scale Orthogonal-IWT (LSOI) features with maximum covariance are then projected into eigen space from where robust training and testing features are selected. For classification of data, Active support vector machine (ASVM) based machine learning technique is used, which generates a less complex procedure compared with traditional support vector machines. ASVM aims to solve a fixed number of linear equations for one-vs-one and one-vs-all multiclass FR. Extensive experiments on Yale, ORL, AR, JAFFE and Georgia-Tech databases have revealed high performance compared with existing FR techniques.
  Keywords: active support vector machine; large scale orthogonal integer wavelet transform; one-vs-one; one-vs-all; multiclass classification.
  
  
• Improving hybrid-layer convolutional neural network system for lung cancer nodule classification using enhanced weight optimisation algorithm  
  by Vikul Pawar, P. Premchand 
  Abstract: In recent times, lung cancer is evolving as a highly life-threatening disease for human beings. According to the WHO, lung cancer disease is the second largest cause of deaths as compared to all other types of cancer. The prevailing available technology is striving to get more exposure in the field of medical science using Computer Assisted Diagnosis (CAD), where image processing is playing a crucial role for detecting the cancerous nodules in computer tomographic images. Augmenting the machine learning techniques with image processing algorithms is becoming a more comprehensive examination of cancer disease in proposed CAD systems. This paper is describes a heuristic approach for lung cancer nodule detection, and the proposed model predominantly consists of the following tasks, which are image enhancement, segmenting ROI (Region of Interest), features extraction, and nodule classification. In pre-processing, primarily the Adaptive Median Filter (AMF) filtering method is applied to eliminate the speckle noise from input CT images of Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): in the LIDC-IDRI dataset, the quality of input image is improved by applying Histogram Equalization (HE) technique with Contrast-Limited Adaptive (CLA) approach. Secondly, in the successive stage the Improved Level-Set (ILS) algorithm is used to segment the ROI. Furthermore, the third step of the projected work is applied to extract the definite learnable texture features and statistical features from the segmented ROI. The extracted features in the subsequent stage of classification are applied to Hybrid-Layer Convolutional Neural Network (HL-CNN) architecture to classify the lung cancer nodule as either benign or malignant. Principally this research is carried out by contributing to each stage of it, where the novel concept of the improved Hybrid-Layer Convolutional Neural Network (HL-CNN) is employed by optimising and selecting the optimal weight using the Enhanced Cat Swarm Optimisation (ECSO) algorithm. The experimental result of the proposed HL-CNN using the weight optimisation algorithm ECSO is achieved an accuracy of 93%, which is comparatively efficient with respect to existing models such as DBN, SVM, CNN, WOA, MFO, and CSO. Moreover, the proposed model conclusively gives a decision on the detected nodule as either benign or malignant.
  Keywords: Computer Assisted Diagnosis (CAD); Computer Vision; Cancer Diagnosis; Image Classification; Image Enhancement; Image Segmentation; Feature Extraction.
  
  
• Lossless EEG data compression using clustering and encoding for fog computing-based IoMT networks  
  by Ali Kadhum Idrees, Marwa Saieed Khlief 
  Abstract: The rapid development in medical devices and communication technologies led to the emergence of the Internet of Medical Things (IoMT), resulting in several new applications that connect to healthcare IT systems through online computer networks. A huge amount of data generated by these applications will be received at the edge gateway periodically to transmit them to the remote cloud for further real-time processing. However, transmitting this huge data to the cloud across the IoT network will place a significant burden on the IoT network. The long processing delays and exchanged data have a considerable influence on the answer time of IoMT applications. The response time of these IoMT applications will be decreased. Therefore, the IoT applications exploit the advantages of fog computing, which is located between the cloud and IoMT devices to minimize the transmitted data and enhance the response time. This paper proposes a Lossless Electroencephalography (EEG) Data Compression (LEDaC) method for fog computing networks based on the IoMT. The LEDaC combines two efficient data reduction techniques: DBSCAN clustering and Huffman encoding, to minimize the volume of IoMT data and then sends them to the cloud for further processing and analysis. The LEDaC works in periods. In each period, the DBSCAN groups the EEG data into groups of similar or identical values. The LEDaC then applies the Huffman encoding to each group, compressing the EEG data indexes with one representative value for each group. The compressed files will be transmitted to the cloud platform, and the original EEG data will be reconstructed without losing any information. The proposed LEDaC method has been tested, and the results show that, in terms of compression ratio, the proposed LEDaC method outperforms the other methods. The LEDaC increased the average compression ratio up to 26.2, 28.6, 27.4, 94.6, 79.5, and 35.4 compared with 2-D SPIHT + AC, JPEG2000, 2-D SPIHT, 1-D SHORTEN, AC, and 1-D SPIHT, respectively.
  Keywords: IoMT; data compression; fog computing; Huffman encoding; DBSCAN clustering.
  
  
• To predict the characteristic impedance of the microstrip transmission line using supervised machine learning regression techniques  
  by Mohammad Ahmad Ansari, Krishnan Rajkumar, Poonam Agarwal 
  Abstract: In this paper, supervised machine learning regression techniques: DNN (Deep Neural Network), SVM (Support Vector Machine) and RF (Random Forest) models have been demonstrated to predict the characteristic impedance of the MTL (Microstrip Transmission Line). Here, MTL input parameters are: width, substrate height and substrate dielectric constant, whereas the characteristics impedance is taken as the output parameter. The dataset is created using MTL analytical formulae. DNN model is developed using Feed-forward Back-propagation learning algorithm, where adam is used as optimizer and relu as the activation function. The regression predictive model of SVM, and RF model of ensemble learning using bagging technique have been developed. It is observed that in DNN model, the minimum MSE is 0.04191 with high execution time, SVM model showed low execution time 0.8327 sec with MSE 0.49. The RF model showed the MSE 0.14 with execution time 1.4296 s.
  Keywords: microstrip transmission line; DNN; support vector machine; random forest; adam; relu; modelling.
  
  
• Terrain identification for intelligent wheelchairs based on geometric properties computation  
  by Yi An, Jianming Ma, Zhuo Sun, Tianqi Han, Yunhao Cui 
  Abstract: With the development of science and technology, intelligent wheelchairs play a key role in the daily life of the elderly. Terrain identification technology plays an important role in the operation of intelligent wheelchairs. In this paper, a typical terrain identification method for road surfaces based on geometric properties computation is proposed. The identification method is divided into four steps. The first step is to collect 3D point cloud data of the terrain in the forward direction through the scanning device carried by the intelligent wheelchair. The second step is to calculate the geometric properties of the point cloud data. The third step is to identify discontinuous points in the point cloud data through the computation results of geometric properties. The fourth step is to identify the terrain according to the type and number of discontinuous points. The identification experiments show that the proposed method has high accuracy.
  Keywords: intelligent wheelchair; discontinuous points; geometric properties; terrain data acquisition; terrain identification.
  
  
• A systems engineering approach for Baxter assistant: programming platform to facilitate the configuration of coBots through natural language  
  by Juan C. Tejada, Alejandro Toro-Ossaba, Juan Berruecos, Santiago Rúa, Daniel Sanin-Villa, Alexandro López-González 
  Abstract: Collaborative robots (cobot) are a great solution for companies that must automate processes without modifying the production line, however, these robots lose flexibility in the application when they need to be located in another point of the production line performing a different task than usual because this action involves programming changes. This paper presents the creation of a programming platform that facilitates the configuration of cobots, to the operator reprogram the robot simply by writing the task that you want to be performed in a web application, without any programming structure, compatible with the flexibility and adaptability to the collaborative robot. To develop this platform, a technology called robotic process automation (RPA) will be used, with which a Bot will be created with the ability to interpret user instructions, structure the corresponding code and program the cobot. Thus, obtaining a platform capable of implementing without the need for additional hardware.
  Keywords: collaborative robots; robotic process automation; natural language; robotic process automation.
  
  
• Diagnosis and management of arrhythmia using machine learning  
  by Pratyaksha Gowda, Chayadevi M L 
  Abstract: Cardiac arrhythmia refers to a variety of heart rhythm disorders in which the heartbeat is irregular, rapid, or sluggish. Arrhythmias come in a variety of forms, some of which have no symptoms. When symptoms are present, palpitations or a sense of a pause between heartbeats may be noticeable. In more extreme instances, lightheadedness, fainting, shortness of breath, or chest discomfort may develop. While most arrhythmias are harmless, some can cause serious complications such as stroke or heart failure. Others might lead to cardiac arrest. Arrhythmia affects millions of individuals throughout the world. Nearly half of all deaths due by cardiovascular disease, or roughly 15% of all deaths globally, are caused by sudden cardiac death. Ventricular arrhythmias account for approximately 80% of sudden cardiac death. Arrhythmias can affect people of any age, although they are more frequent as they get older.
  Keywords: medical imaging; machine learning; arrhythmia diagnosis; KNN; SVM; random forest, decision tree; logistic regression.
  
  
• Slat noise control using active piezo-ceramic actuator  
  by Li Dawei, Kaibo YU, Xiao Wang, CongYun Wang 
  Abstract: An active control method based on piezo-ceramic actuator has been used to reduce pressure fluctuations of shedding vortex on the reattachment point in the slat cavities and sound intensities of far field by simulations and wind tunnel experiments. The periodic active vibration control was imposed on the cusp to change the original properties of shedding vortex on leading edge of slat, and further altered intensities of pressure fluctuations on reattachment point of the vortex shedding. An optimization model has been designed by DDES simulations and measuring experiments, which can reproduce flow field characteristics in the flat of 30P30N airfoil. The wind tunnel experiments show that pressure fluctuations of reattachment point can be reduced by the active control. The fourth peak value of pressure fluctuation of the reattachment point is reduced with the increase of control frequencies, and reached a minimum value when the frequency is 500 Hz. However, the control has little effect on sound intensities of far field.
  Keywords: piezo-ceramic actuators; slat; pressure fluctuation; sound intensities.
  
  
• Optimised reactive resource-aware routing for wireless infrastructure-less networks  
  by Arshad Ahmad Khan Mohammad, Arif Mohammad Abdul 
  Abstract: Wireless infrastructure-less networks comprise mobile devices disseminated in radio communication areas without any central coordinator. The nodes communicate directly if they are within the radio range; otherwise, they rely on other nodes. Thus, nodes should act as a router to forward the information to other nodes. The network permits the nodes to be mobile freely and organises arbitrarily; therefore, any node can participate or leave the network independently. Further, nodes in the network consist of constrained heterogeneous resources. Proper use and management of network resources and characteristics are needed to achieve performance efficiency in communication. The paper designs an optimised reactive resource-aware routing mechanism to use resources effectively and adequately control network characteristics. The proposed mechanism is validated with NS2, and performance is compared with current routing protocols. Results indicate that the proposed mechanism outperforms energy efficiency, packet delivery, and resource use.
  Keywords: infrastructure-less networks; heterogeneous resources; intermediate bottleneck node.
  
  
• FER to FFR: a deep-learning based approach for robust fatigue detection  
  by Rachana Patil, Yogesh Patil, Sheetal Bhandari 
  Abstract: Automatic detection of fatigue from the face provides non-intrusive passive identification of fatigue. The traditional approach of fatigue detection has focused on detecting yawning and eyelid closure. However, fatigue is manifested in the face through the various minute facial features. In this paper, we propose a fatigue detection model, which can learn facial expression features through a deep learning-based facial expression recognition model and provide the same to the fatigue recognition model. Experiments indicate that the proposed approach achieves a qualitative improvement of facial features used for fatigue detection and improves the accuracy quantitatively on the custom Indian fatigue dataset. The approach also allows mitigation of limitations of fatigue datasets of significantly fewer subjects and allows for training fatigue models suitable for unconstrained real-world settings.
  Keywords: deep learning; fatigue detection; vision transformation; fatigue dataset.
  
  
• Exploring profile textual features for cross-network linkability: application to Quora and Twitter users  
  by Youcef Benkhedda, Faical Azouaou, Sofiane Abbar 
  Abstract: Content-based user identity linkage across different social platform has been explored extensively during the last decade. Existing techniques investigated the use of personal discrete attributes such as user name, gender, and email. Using discrete attributes for linking profiles has serious drawbacks, as these attributes are inconsistent and non-authentic in many platforms. In this paper, we suggest a matching approach that explores the textual information contained in user posts rather than their discrete profile information. However, we face major constraints as the profiles textual representation can be very sparse. In addition, the absence of any discrete attributes in the matching scheme makes the problem quadratic, as no profile pairwise filtering can be done. We tackle this by suggesting a clustering method based on locality-sensitive hashing that first clusters users into subgroups based on their topical representation and then selects the true matching pair based on their token signatures. Our solution significantly reduces the problem complexity by reducing the number of pairwise profiles to compare in two orders of magnitudes, and maintains a high matching precision for users that have sufficient amount of textual data.
  Keywords: user matching; identity linkage; network alignment; social networks.
  
  
• Connection between UML use case diagrams and UML class diagrams: A matrix proposal  
  by Bráulio Alturas 
  Abstract: In recent years, the UML language has been one of the most used to conduct information system analysis and design. Being an object-oriented technique, UML provides a vast set of diagrams, in order to represent the various abstractions of the system, of which the most used are the use-case diagram and the class diagram. Often, in the modelling of less complex systems, only these two diagrams are used, where one represents the functionalities of the system (use case diagram) and the other the static structure of the system (class diagram). However, it is often difficult to make the connection between the two diagrams, and mainly, it is difficult to verify when one matches the other. In order to solve this problem, a matrix is proposed that links the two diagrams, using a case study to verify the utility of the matrix.
  Keywords: UML; use case diagram; class Diagram; modelling technique; matrix; information systems; object-oriented; analysis and design.
  
  
• Working on empathy with the use of extended reality scenarios: the Mr UD project  
  by Anna Laska-Lesniewicz, Dorota Kaminska, Grzegorz Zwolinski, Luis Coelho, Rui Raposo, Mário Vairinhos, Eric Haamer 
  Abstract: Empathy has become a central part of Design and is loudly manifested in several frameworks such as universal design, inclusive design or human-centred design. Among a vast group of stakeholders, designers and engineers are the ones who need an expansion of their emphatic horizon. The development of immersive technologies such as virtual or augmented reality may facilitate an individuals empathy development. Extended Reality (XR) merges virtual reality with the real world. Therefore, XR technologies allow us to simulate various physical states, health problems and human body limitations. This paper presents five independent XR scenarios that put potential users in the shoes of people with special needs. Elaborated tasks concern such aspects as vision impairments, autism spectrum disorder, mobility impairments, pregnancy state and some problems of the elderly. All exercises occur in a well-known supermarket environment, as shopping is a usual daily activity for most people. The VR application is prepared for Oculus Quest 2 platform and is supported in some cases by tangible equipment (geriatric suit, pregnancy belly simulator, wheelchair). The proposed simulations were validated by experts who evaluated the quality of the proposed tasks and the possibility of simulating selected limitations or issues in XR. Ongoing development and testing of the XR application will provide further in-depth views on its usefulness, acceptance and impact in increasing empathy towards the challenges faced by the personas portrayed.
  Keywords: virtual reality; extended reality; immersion; empathy; disability simulation; presence; learning; education; HMD.
  DOI: 10.1504/IJCAT.2022.10054963
   
  
• An automatic image retrieval system using multi-scale local ternary pattern  
  by Megha Agarwal 
  Abstract: Content based image retrieval (CBIR) addresses the issue of finding out the relevant images automatically from the vast image repositories. Images are compared based on the extracted features and, hence, the feature extraction is highly responsible in CBIR system performance. In this paper, a unique feature, multi-scale local ternary pattern (MSLTP) is designed. Mostly, local patterns considers original image for feature extraction but in MSLTP, the images are analysed in five different scales and, hence, all the image information whether, fine or coarse is captured in feature extraction. Along with this, texture information is computed by taking small neighbourhoods and comparing surrounding pixels. The pattern of intensity variation in local neighbourhoods is captured. Performance of the system is validated on very distinct benchmark datasets Corel 1k and MIT VisTex. Significant improvement is observed in terms of retrieval precision and recall, as compared to the other handcrafted features available in the literature.
  Keywords: image retrieval; local pattern; Corel 1k; MIT texture.
  
  
• A hybrid convolutional neural network model for detection of diabetic retinopathy  
  by Musa Alshawabkeh, Mohammad Hashem Ryalat, Osama Dorgham, Khalid Alkharabsheh, Mohammad Hjouj Btoush, Mamoun Alazab 
  Abstract: Diabetic retinopathy (DR) causes vision loss. Regular eye screening has to be done to provide the appropriate treatment and for vision loss prevention. Globally, patients with DR are increasing, which leads to work pressure on specialists and equipment. Fundus images are a key factor in effective retinal diagnosis. In this paper, a deep-learning approach is proposed to detect DR from retinal images. The proposed approach involves a combination of four effective techniques: image augmentation, contrast limited adaptive histogram equalization, CNN, and transfer learning and ensemble classification. The results show the proposed approach obtained high values of accuracy (93%), precision (95%), and recall (96%), and more stability compared with other approaches.
  Keywords: deep learning; diabetic retinopathy; eye diseases; retinal diagnosis; retinal images; convolutional neural networks; medical applications; ensemble classification.
  
  
• Prediction model for total amount of coke oven gas generation based on FCM-RBF  
  by Lili Feng, Jun Peng, Zhaojun Huang 
  Abstract: The rational use of Coke Oven Gas (COG) is of great significance to improve the economic efficiency of enterprises. In this paper, a COG generation prediction model based on fuzzy C-mean clustering (FCM) and radial basis function (RBF) neural network is proposed to address the problems such as the difficulty of accurate modelling of COG generation process and the difficulty of real-time flow prediction. Firstly, the coke oven production process is analysed and correlation analysis is used to select the influencing factors. Secondly, the FCM is used to classify the working conditions of the coke oven, and the appropriate number of working conditions is selected through experiments. Finally, the prediction models under different working conditions are established separately by using RBF. The experiments were carried out using actual industrial production data, and the experimental results showed that the model could provide guidance reference for the dispatchers.
  Keywords: coking oven process; fuzzy C-means clustering; prediction model; radial basis function neural network.
  
  
• Efficient load balancing in cloud computing using HHO improved by differential perturbed velocity and TEO  
  by Uttam Kumar Jena, Manas Ranjan Kabat, Pradipta Kumar Das 
  Abstract: Load balancing is one of the primary aspects of cloud computing to avoid situations of being overloaded or underloaded in the node. This paper aims to carry out the dynamic load balancing of non-determent independent tasks in the cloud network and resolved through the hybridisation of an improved version of the Harris Hawk Optimisation Algorithm (HHO) improved by differential perturbed velocity and Thermal Exchange optimization (TEO). The main motivation of hybridising is to intensify the diversification ability of the device through the load balance with the VMs, in order to optimise different matrices and enhance the convergence speed. The strength of the algorithm has been authenticated by relating the outcome gained from simulation and real platform processes with the surviving load balancing. The conclusions drawn from the simulation and comparison results illustrate that the projected procedure is outstripping its opponent in the manner of different matrices.
  Keywords: load balancing; throughput time; cloud network; optimisation.
  
  
• Ensemble-based software fault prediction with two-stage data pre-processing  
  by Shubham P. Kulkarni, Sanjeev Patel 
  Abstract: Software fault prediction is the process of identifying the software modules which are more likely to be defective or faulty before the testing phase of software development life-cycle model. We use software metric values of different modules for the known software project to train the software fault prediction model. Our objective is to implement the ensemble-based models on software fault datasets along with feature selection and data re-sampling techniques to achieve the improved performance. In this paper, we have designed a two-stage data pre-processing technique on the dataset before passing it through the ensemble-based model for training. It has been found that the two-stage pre-processing model outperforms the general ensemble-based model. It gives an improvement of 1% to 6% for all the used classifiers viz., Bagging, Dagging, Rotation Forest, Random Forest, and AdaBoost.
  Keywords: software fault prediction; ensemble-based model; SMOTE; feature selection.
  
  
• Group Delay-based Minimum Variance Distortion-less Response (MVDR) Cepstral Features for Speaker Identification in Whispered Speech
  by Vijay Sardar, Manisha L. Jadhav, Saurabh H. Deshmukh, Makarand M. Jadhav 
  Abstract: The whispering voice shows a wide difference in characteristics compared to the neutral voice. It makes identification of a person from the whispered sound difficult. The Group delay function (GDF) in its spectral form considers the phase information in the short-time FT phase function, which is otherwise ignored in traditional front-end processing. A minimum variance distortion-less response (MVDR) based on smoothing on the denominator of group delay that enhances speech quality and intelligibility is proposed in this paper. The experiment uses MVDR spectral coefficient features with a multi-class Support Vector Machine (SVM) for classification. The proposed method reported an improvement of 2.41 % over the baseline system using the CHAINs database and SVM classifier. The five-fold cross-validation is exercised for accuracy and speaker error rate (SER) to verify the consistency of the results. The proposed system is also evaluated for False Positives (FP) and Precision and reported the enhancement compared to the baseline system.
  Keywords: Whispered speech; Group delay function; MVDR; Support Vector Machine; MFCC.
  
  

Special Issue on: ICIGP 2022 Advances in Image and Graphics Processing

• MGU-Net: a multiscale gate attention encoder-decoder network for medical image segmentation  
  by Le Liu, Tao Lei, Qi Chen, Jian Su, Yong Wan, XiaoGang Du 
  Abstract: Medical image segmentation-the prerequisite of numerous clinical needs-has been significantly prospered by recent advances in encoder-decoder networks. However, uneven reflection of human organs and the subjects tremor and movement cause blurred edges in the image, which is difficult to segment, and need more details and context information to release this problem. Most of the existing Unet-like architectures do not take into account the multiscale characteristics of medical images and make full use of the spatial information and channel information of feature maps, resulting in the loss of detail information. This paper proposes a multiscale gate attention (MGU-Net) encoder-decoder network. Firstly, we use multiscale blocks to focus on the fusion of contextual information. Besides, we use two gate attention to deploy more detailed information. On three different public datasets, compared with other State-Of-The-Art (SOTA) methods, the proposed method achieves an improvement.
  Keywords: medical image segmentation; gate mechanism; multiscale feature fusion.
  
  
• Efficient adaptive rendering of planetary-scale terrains  
  by Zafar Masood, Zheng Jiangbin, Muhammad Irfan, Idrees Ahmad 
  Abstract: Planetary-scale terrain requires adaptive simplification and tessellation for high-performance rendering. Real-world visualisation requires adaptive rendering of the world-scale geometric model, which is a challenging task especially on consumer-level notebooks. In this work, we propose a method for the adaptive rendering of the earth's surface model using modern GPU features. The multi-resolution model is partitioned into patches and are submitted efficiently using the geometry-instancing feature for high-performance rendering. Patch-based algorithm culls world-scale model in graphics pipeline to reduce simplification and rendering load. The proposed method performs the adaptive simplification of patches with geometric-error control in image space for optimum triangulation. Simplified patches are tessellated using power-of-two tessellation factors to avoid boundary cracks. After tessellation, a three-dimensional surface is constructed by displacement mapping using a regular height data grid. Evaluation of the method is performed and achieved a stable high frame rate without any stuttering. Comparative evaluation of the proposed method with clipmaps and hardware-based tessellation methods is performed and results are presented.
  Keywords: real-time graphics; level of detail; adaptive simplification; hardware acceleration; geographic information systems.
  
  
• Image super-resolution algorithm based on V-transform combined with neural network  
  by Nan Nan, Shijie Yue, Ruipeng Gang, Chenghua Li, Ruixia Song 
  Abstract: Single image super-resolution aims to increase the visual quality and size of low-resolution images. Although the existing deep learning-based methods have achieved promising results, there is still room for improvement in both subjective and objective effects due to the inability of deep convolution to balance low-frequency content and high-frequency details in the reconstruction of complex scenes. To solve this problem, we propose a V-transform based image super resolution model combined with convolution(VTSR). The VTSR is mainly composed of three parts: the V-transform block, feature fusion model and the upsampling module. Test experiments on four standard datasets show that our proposed V-transform based image super resolution model combined with convolution can achieve better results than most methods at all scales. The three innovations we propose have positive effects on super-resolution tasks to varying degrees.
  Keywords: single image super-resolution; CNN; V-transform; frequency domain.
  
  
• Dynamic visualisation and measurement of cartilage morphology by magnetic resonance imaging-based knee kinematics  
  by Wei Quan, Zhe Wang, Lik-Kwan Shark 
  Abstract: To address the limitation in static imaging for clinical diagnosis of knee joints based on a snapshot of the knee in a fix pose, an approach is presented for quantitative assessment of knee joints in a dynamic manner. The core of the proposed approach is based on kinematics of the knee bones, whereby articulation of knee joints derived from magnetic resonance imaging is emulated by capturing and mimicking the movement of an artificial anatomical knee model. Through bone-based kinematic emulation, dynamic visualisation and measurement of cartilage morphology are demonstrated by focusing on tibiofemoral cartilage thickness as a function of knee joint movement angle. In particular, the differences in dynamic tibiofemoral cartilage thickness between two knees with mild and severe osteoarthritis are illustrated to show the effectiveness and potential of the proposed approach. Also presented is an interactive visualisation and measurement tool based on Matlab for dynamic knee joint assessment.
  Keywords: kinematics; magnetic resonance imaging; knee assessment; motion tracking; modelling and emulation; knee joint landmark; cartilage thickness.
  
  
• Effsemble: faster, smaller, and more accurate ensemble networks for thoracic disease classification  
  by Arren Matthew Antioquia 
  Abstract: Convolutional Neural Networks (CNNs) are being adapted to various computer-aided diagnosis applications, including recognising thoracic diseases. To improve classification performance, recent solutions alter the structure of existing networks or require additional prior information for training. Other approaches demand massive computational requirement to increase classification accuracy. In this paper, we propose a family of efficient ensemble networks called Effsemble to accurately recognise thoracic diseases without additional layers, extra input data, nor large computational overhead. Our proposed approach achieves the highest average AUROC score of 80.04% in the ChestX-ray14 dataset. Moreover, our Effsemble-2 outperforms other state-of-the-art methods, while decreasing the parameter count by 31.7M and increasing inference speed by 2.75
  Keywords: thoracic disease classification; image classification; convolutional neural networks; ensemble learning; deep learning.
  
  

Special Issue on: Edge Computing and Artificial Intelligence Driven Technologies for Education Improvement

• IoT for smart English education: AI-based personalised learning resource recommendation algorithm  
  by Fang Wang 
  Abstract: The development of smart English education provides talent support for national economic and social development, solves the difficult problems in current education, and enhances national competence. At present, the Internet of Things (IoT) is constant maturing of continuous development of information technology, providing a solid technical foundation for the development of smart English education. IoT mainly senses and collects data through multi-sensors, and uses a variety of communication technologies to establish a network to realise the connection between things and things.
  Keywords: smart English education; IoT; AI; sensor; recommendation model.
  
  
• Sports injury detection mechanism based on multi-sensor fusion  
  by Zun Liu, Xue Han, Yi Yang, Wei Wang, Fuquan Liu 
  Abstract: Cardiovascular disease and muscle damage caused by human sports injuries are increasing. Therefore, it is of great research significance to greatly reduce the damage to the human body through sports injury monitoring, detecting the degree of injury, and taking effective measures in time. In this paper, we propose a sports injury detection mechanism based on multi-sensor fusion. First, we determine the motion area of the human body by computing the cumulative difference between frames. Second, we use a wavelet method and an AutoEncoder model to design two feature extraction methods. Third, we combine the extracted features to detect the sports injuries. The experimental results verify the effectiveness of our proposed method.
  Keywords: sports injury detection; multi-sensor fusion; neural network; feature extraction.
  
  
• A novel LightGBM-based industrial internet intrusion detection method  
  by Zhiqiang Lv 
  Abstract: This paper proposes an active learning-based intrusion detection system. The system introduces expert annotation into the intrusion detection process, and combines the active learning query strategy with LightGBM to solve the problem of low accuracy of the intrusion detection system when the training samples are scarce. First, the process of data preprocessing is applied. Features are extracted from the traffic, and the borderline SMOTE method is introduced to improve the samples distribution. Then, the LightGBM algorithm is adopted for feature selection to reduce the data dimension. Next, the most valuable training samples are selected and labelled by human experts. The training samples are then fed into the classifier, while the Bayesian optimisation is applied to optimise the hyperparameters of the classification model. Finally, a set of experiments are performed to evaluate the performance of our method.
  Keywords: industrial internet; intrusion detection; active learning; LightGBM.
  
  
• A novel neural network-based 3D animation model classification method  
  by Ximan Shi 
  Abstract: The rapid development of information technology has also brought new vitality to art design. The 3D animation model making is a new multimedia technology based on computer technology. In order to efficiently organise and use the 3D model resources, researchers focus on how to achieve effective retrieval and classification. In order to realise the recognition and classification of 3D models, a novel network model called 3DSmallPCapsNet is proposed in this paper, based on the feature that Capsule Network (CapsNet) exploits vector neurons to store feature space information. The proposed method can extract more representative features while reducing the model complexity. To evaluate our method, three different methods, which are MeshNet, Shape-DNA and GPS-embedding, are compared. The experimental results on datasets SHREC10 and SHREC15 show that the proposed method has better performance.
  Keywords: 3D model classification; capsule network; pooling; animation model.
  
  
• An edge computing-based evaluation and optimisation of an online higher vocational education mechanism  
  by Jingtang Jia 
  Abstract: As a skill education and employment education, higher vocational education must comprehensively improve students' various skills, so that they can fully adapt to the social economic development and the actual needs of the post after graduation. However, the outbreak of COVID-19 disrupted students' normal life, making online teaching the main teaching mode.
  Keywords: edge computing; higher vocational education; task offloading; particle swarm optimisation.
  
  
• Multidimensional meteorological data analysis based on machine learning  
  by Jianxin Wang, Geng Li 
  Abstract: Multidimensional meteorological data has very important application scenarios, and how to effectively analyse and use it is a challenging problem. This paper proposes a multi-dimensional meteorological data analysis method based on an improved Bayesian neural network. This paper considers the example of wind power forecasting for wind farms. The input data can be divided into two categories, which are the multidimensional meteorological data and historical data of wind power. First, the raw multidimensional meteorological data are preprocessed using principal component analysis (PCA). Then, the processed meteorological data and historical wind power data are fed through the long- and short-term memory network to achieve data feature extraction and further data dimensionality reduction. At last, they are input to the improved Bayesian neural network to achieve data fitting. This paper selects the data of 12 wind farms in a certain region of China for simulation experiments. Our proposed method is compared with BP-neural network and support vector machine to evaluate its performance. The experimental results show that the method proposed in this paper has good performance.
  Keywords: meteorological data; multidimensional data; Bayesian neural network; wind power forecast.
  
  
• Energy-saving smart city: an edge computing-based renovation and upgrading scheme for old residential areas  
  by Zhi Zhao 
  Abstract: The renovation of old communities has become an important issue in the current development of new urbanisation. The development of edge computing provides a powerful pillar for the energy-saving renovation of old residential areas. Accurately predicting the electricity usage can provide a more personalised electricity consumption plan for the users in the community, thus making the overall energy saving possible. Therefore, we propose a power prediction model based on the stacking model to provide a strategy for saving power and energy in old communities. First, we adopt the Word2vec algorithm to extract the discrete feature word vector and to capture the co-occurrence relationship from the discrete feature. Second, we adopt a neural network model to perform feature extraction on for continuous features. Third, we design a power prediction model based on the stacking model by using XGBoost algorithm, LightGBM algorithm, and linear regression. The experimental results prove that the method proposed in this paper has good prediction performance.
  Keywords: energy-saving; smart city; neural network; feature extraction.
  
  
• A novel deep learning driven robot path planning strategy: Q-learning approach  
  by Junli Hu 
  Abstract: As the basis of mobile navigation technology, path planning has attracted the attention of the majority of scholars.
  Keywords: robot; path planning; deep learning; DQN; memory.
  
  
• Machine learning for English teaching: a novel evaluation method  
  by Yang Yang 
  Abstract: This paper proposes a novel oral English scoring system based on machine learning. The system can be deployed on the end side (mobile app) through the internet and can be used to assist teachers in evaluating students' oral English pronunciation, fluency and the tunnel degree. An attention-based LSTM (Long Short-Term Memory) neural network is employed in the paper, which can process and analyse speech signals effectively. Meantime, a large amount of training data is collected for network training. We compare the novel oral English evaluation system and the experts evaluation results. The verification results show that the oral English evaluation system based on machine learning not only can achieve the ability of the English experts, but also has higher accuracy and can identify more oral pronunciation problems.
  Keywords: college English teaching; oral English; machine Learning; LSTM; attention.
  
  
• Ideological and political empowering English teaching: ideological education based on artificial intelligence in classroom emotion recognition  
  by Liqun Zhang 
  Abstract: Emotions play an important role in human decision-making, interaction and cognition. Accurately identifying human emotions can provide effective support for human decision-making and solutions. Reflecting the learning situation with classroom emotions can assist teachers in implementing teaching interventions, so as to help teachers carry out accurate teaching. In this paper, we propose a classroom emotion recognition method based on multi-modal fusion of speech and text. We first adopt the CNN and LSTM to extract the spatio-temporal feature from the speech data. Then, we adopt a LSTM model to perform feature extraction on text data. After obtaining these two types of feature, we design a fusion model based on attention mechanism. The experimental results prove that the method proposed in this paper has good prediction performance, which is of great significance for the classroom emotion recognition.
  Keywords: emotion classification; multi-modal feature fusion; neural network; feature extraction.
  
  
• Metaverse-enabled fine art appreciation: an aesthetic based on visual expression  
  by Ying Guo 
  Abstract: Based on the idea of the metaverse, this paper adopts an aesthetic way of visual expression to appreciate art, that is, transforming a two-dimensional (2D) image to a three-dimensional (3D) image.
  Keywords: metaverse; art appreciation; visual expression; 3D; hole filling.
  
  
• AI-based artist style appreciation: folk art in the Central Plains oriented implementation platform  
  by Wei Tian 
  Abstract: The folk art in the Central Plains of China is rich in variety and has strong vitality. The culture of the Central Plains has a long history, and folk art is considered to be a representative type of folk culture in the Central Plains. Given this, this study proposes an image of folk art in the Central Plains style transfer algorithm with salient region reservation by fast style transfer algorithm. By introducing saliency detection network to generate saliency map of composite image and content image, which is helpful to improve the quality of stylised image. Experiments show that the stylized image generated by the algorithm proposed in this paper not only has satisfactory colour and texture, but also reserves salient regions in the content image, which is conducive to the appreciation of the folk art in the Central Plains.
  Keywords: folk art; Central Plains of China; artificial intelligence; image style transfer; stylised image.
  
  

Special Issue on: ICIGP 2022 Advances in Image and Graphics Processing

• MERCoL: video-based facial micro-expression recognition via bimodal contrastive learning  
  by Yanxin Song, Pengyu Wang, Hao Sun, Lei Chen, Xianye Ben 
  Abstract: Micro-expressions are short-duration, subtle and involuntary facial gestures that usually reveal people's genuine mental activity and have many real-world applications. Owing to the transient nature, low intensity and difficulty of capturing micro-expressions, many micro-expression recognition algorithms based on handcrafted features and deep learning methods are not accurate enough. In order to simultaneously extract the common and distinctive features of different micro-expressions in limited dataset samples, we propose a micro-expressions recognition framework based on bimodal contrastive learning called textbf{MERCoL}. Specifically, the network mainly includes three modules: bimodal feature extraction module, bimodal contrastive learning fusion module and classification module. First, the micro-expressions sequence is divided into RGB and optical flow sequence, and the loss function between them is constructed by contrastive learning so that the network can learn common bimodal features. Second, bimodal features are fused and labelled data is used to optimise the network to learn distinctive features. At last, we conduct broad experiments on CASME II, SAMM and MMEW datasets and demonstrate the superiority of our algorithm compared with other state-of-the-art methods.
  Keywords: micro-expression recognition; contrastive representative learning; optical flow.