International Journal of Reasoning-based Intelligent Systems (47 papers in press)
Parallel K-Means Algorithm based on Two Stage Clustering of Large Data
by Xia Wendong, Liu Yuanfeng, Chen Deli
Abstract: Aiming at the fact that the algorithm communication time occupation ratio is too high and the practical application value is limited under the Mapreduce mechanism, a Hadoop-based two stage parallel c-Means clustering algorithm is proposed to solve the classification problem of super large data. First, the Hadoop-based two stage parallel fuzzy c-Means clustering algorithm is proposed to process the clustering of large data; and a protocol-based group typical individual reduction strategy is used to improve the time complexity of the MPI communication model of Mapreduce, so as to improve the overall efficiency of the algorithm; secondly, the interference of bad data items can be effectively eliminated by the selective group reduction algorithm, so that the algorithm in this paper has higher operating efficiency and clustering success rate. In terms of parallel rate and speedup ratio, the parallel rate and speedup ratio of the proposed algorithm in this paper on the large data set is better than the performance of the small data set, which means that the algorithm in this paper can adjust itself according to the amount of data in real time. The simulation results show that the performance of PGR-PFCM algorithm is better in the processing of large data.
Keywords: parallel algorithm; fuzzy clustering; K-Means; big data; two stages.
Glossary Applications Model for Financial Terms with Boyer-Moore-Horspool Method Based on Mobile Application
by Nazori Agani, Arfian Maulidan
Abstract: String Search or the string matching is one aspect that is very important in terms of data processing, in addition the problem string matching is also one of the problems that are well known in the world of informatics. Some examples of the implementation of the string matching problem is in matching a string in a text editor application such as Microsoft Word, or in the case of bigger, ie matching website by entering key words as it has been implemented on search engines such as Google Inc. The general process in the searching a string is looking for a string that consists of some of the characters (called pattern) in a large amount of text. Search string is also used to look for patterns of bits in a large number of binary files. Problems begin to arise if the search process occurs in a lot of data and complex, of course, this would be very time-consuming and resources owned, so the search technique effectively and efficiently will be needed. The purpose of this study is to utilize a String Matching algorithm that is by using the Boyer-Moore-Horspool in the search data is a list of financial terms. Boyer-Moore algorithm-Horspool used to search for any string precision (word or phrase) entered with a high degree of accuracy of search, because it uses pattern matching from right to left. In addition to the HTML5-based application model, application model matching string can later be run multiplatform not only on mobile devices, but also be able to run on a desktop browser.
Keywords: String Matching; Search Engine; Boyer-Moore-Horspool; Mobile; Browser.
A File Sharing System in Peer-to-Peer Network by a Nearness-Sensible Method
by Vimal .S, Srivatsa S.K
Abstract: For a comprehensive Peer-to-Peer file sharing system dynamic file query is substantial where its performance can be enhanced by clustering of peer that can also considerably improve the efficiency. Depending upon physical nearness and peer interest peers are clustered in current work. File replication algorithm has been employed that creates replicas for the requested file to enhance the efficiency. Compared to unstructured P2P the efficiency is high for structured P2P which is difficult to analyze because of their rigid topology. We have developed Nearness and Interested Cluster (NIC) super peer network to improve the efficiency of file location in current years for P2P system but few works rely on peer interest and physical nearness. Various methods have been used to improve intra-sub-cluster searching. Here the interest is categorized into sub-interest then they are linked according to common-interest. File searching delay is minimized where an overlay is built to link. Flower filter is employed to enhance the efficiency and reduce the overhead. On comparing Nearness Sensible I-clustered System with other system the efficiency has been traced. The effect of enhancing the efficiency using intra-sub-cluster searching is observed in experimental results.
Keywords: File Replication; Peer to Peer Networks; Flower Filter mechanism.
Temperature Aware Power Optimization based 8-bit MAC Architecture for Low Power DSP Applications
by Haripriya D, Govindaraju C, Sumathi M
Abstract: Temperature aware adaptive voltage scaling based low power 8 bit Multiplier-Accumulator (MAC) architecture for Digital Signal Processing (DSP) has been presented in this paper. Temperature increases dynamic power, static power and leakage power dissipation in the electronic circuits, hence it is mandatory to construct a circuit which minimizes the dynamic power, static power and leakage power adaptively according to the current temperature so that the performance of the overall system is not degraded much. The proposed temperature aware adaptive voltage scaling is very effective method to minimize the dynamic power, static power and leakage power consumption without degrading the performance of the system. The dynamic power consumed by the conventional MAC is 3.441mW when the temperature is 150
Keywords: Adaptive dynamic voltage scaling; Digital signal processing; Dynamic power;; Leakage power; Multiply and accumulate; Static power; Temperature aware.
Data Mining and Economic Forecasting in DW-based Economical Decision Support System
by Min Zhang, Rui Qi
Abstract: Decision demand has hierarchies for different users and the decision analysis demand in various area and field have particularity according to different topics. Since traditional MIS is hard to meet the demand of analysis and processing of growing mass data, a novel decision support system(DSS) is urgent to be proposed for decision makers. Based on data warehouse, data mining and OLAP technology, we propose a DSS with modular design, and explain the structure and key technologies of it in this article. Our study establishes multidimensional data-set for OLAP analysis to perform slicing, dicing, drilling and rotation operation. In data mining, for the problems of large data-set such as long learning time and decreasing generalization ability, an SVM accelerating algorithm based on boundary sample selection is put forward. The system test results demonstrate that the data mining has better prediction effects on economical forecasting. Therefore, the research has better practicability and higher accuracy, which shows certain value of popularization and implementation.
Keywords: data mining; data warehouse; DSS; OLAP; SVM; Economic forecasting.
Toward an Automatic Summarization of Arabic Text Depending on Rhetorical Relations
by Samira Lagrini, Nabiha Azizi, Mohammed Redjimi, Monther Al Dwairi
Abstract: Rhetorical relations between two text segments are crucial information and have been proven useful for many natural language processing (NLP) applications. In this paper, we propose a supervised approach for automatic identifying of rhetorical relations in Arabic texts. To the best of our knowledge, this is the first model that attempts to identify both implicit and explicit rhetorical relations between elementary discourse units within the rhetorical structure theory (RST). To carry out this research, we developed a discourse annotated corpus following the RST framework with high reliability.Relations annotation was done using a set of 23 fine-grained relations enriched with nuclearity annotation. To automatically learn these relations, we reuse some state of the arts featuresand contribute new lexical and semantics' features. The experimental results, on fine-grained and coarse-grained relations, showed that our model achieved the best performance relative to all baselines
Keywords: Rhetorical relations; Arabic language; Rhetorical structure theory.
MGA-TSP: Modernized Genetic Algorithm for the Traveling Salesman Problem
by Ra'ed M. Al-Khatib, Mohammed Azmi Al-Betar, Mohammed A. Awadallah, Khalid M. O. Nahar, Mohammed M. Abu Shquier, Ahmad M. Manasrah, Ahmad Bany Doumi
Abstract: This paper proposes a new enhanced algorithm called Modernized Genetic Algorithm for solving the Traveling Salesman Problem (MGA-TSP). Recently, the most successful evolutionary algorithm used to solve the TSP problem, is the GA algorithm. One of the main obstacles for GA is building its initial population. When initiating the GA with a strong initial population, the convergence rate and the diversity aspect will be more stronger. Therefore, in this paper, a new local search mechanism based on three neighborhood structure operators (Inverse, Insert, and Swap) along with 2-opt is utilized. This adapted neighborhood structure operators are employed to generate the initial population for GA algorithm. In addition to building powerful initial population for TSP, the main operators (i.e., crossover and mutation) of GA during the generation process should be also enhanced for TSP. Therefore, the recent and powerful crossover operator called EAX is utilized in the proposed MGA-TSP to enhance its convergence behavior. In order to validate the performance of the proposed algorithm we used TSP datasets, have different complexities and sizes. The sizes of the dataset-cities, range from 150 to 33810 cities. Initially, the impact of each neighboring operator on the performance of the proposed algorithm is studied. In conclusion, our proposed method achieved the best results. For comparative evaluation, the results obtained from our proposed MGA-TSP method is compared with those obtained by six well-regard methods using the same TSP instances. The proposed method is able to outperform other comparative methods in almost all TSP instances used.
Keywords: Traveling Salesman Problem; Optimization; Genetic Algorithm; Neighboring Operators.
Version.01: Design and Development soft actuator prototype for Surgical Lighting System
by Sandesh Ghate, Guntis Kulikovskis
Abstract: Surgical luminaires are used for illumination of wounds during surgery. For optimal illumination surgical luminaries need to change their orientation several times during surgery. The aim this project is simplify and optimize the design of the surgical lighting system to overcome the structural limitation and to reduce singularity. Surgical lighting system is made up of 6 links and in order to achieve 5 Degree of freedom. This is an attempt to reduce number of linkages in SLS by introducing a bendable soft actuator. Pneumatic bending actuator made of silicone rubber undergoes the desired deformation when each chamber is pressurized. Soft actuator because of flexibility provides advantage to overcome the mechanical singularity faced by existing surgical lighting system. Mathematical model based on geometric deformation has been presented. A fourth degree polynomial approximation has been used for characterize behavior of each chamber of actuator.
Keywords: Surgical Lighting System; Soft Actuator; Surgical Luminaries. Pneumatic Bending Actuator.
Intelligent control algorithm for USV with input saturation based on RBF network compensation
by Renqiang Wang, Jianming Sun, Hua Deng, Keyin Miao, Yue Zhao
Abstract: A type of intelligent control algorithm of course tracking for USV was proposed on the basis of RBF network approximation and compensation with input saturation. First of all, sliding surfaces with integrator were designed with the sliding mode variable structure control technology. Then, radial basis function neural network was applied to approximate compensate the system input saturation. Furthermore, second-order system observer was introduced to overcome the bounded outside interference. Finally, the control algorithm for USV was
Keywords: USV motion control; intelligent control; RBF neural network; sliding mode control; saturation.
Application research of improved ICA algorithm for initial population establishment based on optimisation goal in limited-buffer flexible flow shop scheduling problem
by Yongqing Jiang, Bin Duan
Abstract: To solve the limited-buffer flexible flow shop scheduling problem (LBFFSP), the LBFFSPs mathematical model is established, and an improved imperialist competitive algorithm (IICA) is proposed as the global optimising algorithm, which contains three modifications including the discretisation processing, reform operation and the elite-individual retention strategy compared with standard imperialist competitive algorithm. In order to further improve algorithms efficiency for searching the optimal solution, the initial population establishment method based on optimisation goal is designed. In addition, the individual selection mechanism on the basis of hamming distance is applied to improve the quality of initial solution in the initial population. The algorithm parameters are analysed to determine the optimum parameter values by designing the simulation experiments. Finally, the effectiveness of the improved imperialist competitive algorithm (IICA) in solving the limited-buffer flexible flow shop scheduling problems is verified in comparison with other algorithms through example test.
Keywords: limited-buffer; improved imperialist competitive algorithm; IICA; initial population establishment; hamming distance.
Optimizing the Mining Strategy of Web Page Based on Ant Colony Algorithm of Information Entropy
by Meiwen Guo, Jianping Peng, Yuanping Zhang, Junxiong Zhao, Liang Wu
Abstract: The speed and quality for browsers to obtain page information are determined by the accuracy degree of web page information filtering. This research improved ant colony algorithm, introducing the information entropy with the ability to judge the probability of occurrence of information and adjusting its operation order. The study uses Sina homepage information from January 2017 to August as a sample, Four indexes are used to evaluate the improved algorithm, which are maximum iterations, average execution time, average error rate and error percentage. It is found that the four indexes of improved algorithm has better effect on the precision of information mining than before, and the cost of this method has not increased significantly. This algorithm is used to provide web page information layout as well as information placement strategies, so as to help website operators and web page designers to further enhance the design and operation efficiency.
Keywords: Data Mining，Ant Colony Algorithm，Information Entropy.
Island-based whale optimization algorithm for continuous optimization problems
by Bilal Abed-alguni, Ahmad Klaib, Khalid Nahar
Abstract: The whale optimization algorithm (WOA) is a newly proposed evolutionaryrnalgorithm that uses a simulation model based on the bubble-net hunting mechanism of humpback whales to find solutions for different classes of optimization problems. WOA may occasionally converge to suboptimal solutions because of the loss of diversity in its population of candidate solutions. The island model is a distributed approach that is commonly used to control the population diversity in evolutionary algorithms. This paper introduces an improved version of WOA namely island-based whale optimization algorithm (iWOA) that incorporates the island model into WOA. In iWOA, the populationrnof candidate solutions is divided into separate sub-populations called islands. The improvement loop of WOA is then applied separately to the candidate solutions in each island. After a predetermined number of generations, a number of candidate solutions are swapped between islands through a process known as migration that is based on the random ring topology. The migration process is conducted to maintain the diversityrnof population and also to allow each island to exchange candidate solutions with a selected neighbouring island. The iWOA algorithm was tested and compared to well-known optimization algorithms using 18 standard benchmark functions. The simulation results indicate that iWOA improves the accuracy of results compared to WOA and other popular evolutionary algorithms.
Keywords: Whale Optimization; Island model; Structured population; Optimization; Evolutionary algorithm.
Automatic absence seizure detection and early detection system using CRNN-SVM
by Niha Kamal Basha, Aisha Banu Wahab
Abstract: In this paper the new model is proposed to automatically detect and predict absence seizure using hybrid deep learning algorithm [convolutional recurrent neural network (CRNN)] with single channel electroencephalography (EEG) only as input. This model comprises of four steps: 1) single channel segmentation process; 2) extraction of relevant features using convolution network; 3) recurrent network for detection and early detection; 4) SVM have been used as last layer to obtain a result with respect to time. This model enhances the feature extraction by feeding the raw input into convolutional layer, improves the detection with gated recurrent unit (GRU) and reduces the early detection rate with support vector machine (SVM). Our proposed model achieves 100% overall accuracy on seizure detection as normal and absence seizure and detect within three seconds of the overall seizure duration. Also this model can be act as a generic model for classification task with detection and early detection of bio-signal (EEG, ECG and EMG).
Keywords: absence seizure; convolutional recurrent neural network; RNN; electroencephalography; gated recurrent unit; GRU; normal and ictal subject; rhythmic frequency; seizure detection; early detection; sampling rate; support vector machine; SVM; statistical features.
Hybrid Neural Network with Bat Approach for Smart Grid Fault Location
by Mangal Dhend, Rajan Chile
Abstract: Abstract: This paper proposes identification of fault location in smart distribution grid based on artificial intelligence using currents and voltages; measured with the help of sensor nodes in distribution system. The approach presented here is the hybrid bat algorithm with neural network, implemented on latest smart distribution system which comprises distributed generation. The fault lengths for various types of faults on distribution feeders are recognized using system parameters, measured before and after the occurrence of a fault. For verifying the performance of proposed algorithm, the MATLAB based coding is developed and executed on sample modified IEEE test feeders. The performance of proposed technique is compared with the simple neural network method. The proposed method founds more accurate and fast in speed.
Keywords: Artificial neural network; bat approach; fault location; smart grid; distribution system.
Uyghur short-text classification based on reliable sub-word morphology
by Sardar Parhat, Mijit Ablimit, Askar Hamdulla
Abstract: In this paper, we research some short-text classification methods for a low resource language combined with reliable stemming and term extraction methods. Uyghur is a morphologically rich agglutinative language in which words are formed by a stem attached by several suffixes, and this property causes infinite vocabulary in theory. As the stems are the semantic entities, stem based text classification is the promising way for the low resource morphologically derivative languages. And it is also an efficient way in NLP to extract and predict out-of-vocabulary (OOV) and misspellings based on context information. The word(or stem)-vector based morphological analysis incorporating stem-vector to text classification is a novel approach for the Uyghur language. Our stemming method extracts noisy stems robustly and decrease the particle lexicon to 1/3 of word lexicon and improve the coverage, thus suited for small corpora with high OOV rate resources. And the highest accuracy of 93.5% is obtained in 9 categories of short texts based on stem-vector with CHI-2() feature.
Keywords: Word embedding; text classification; morphology; Uyghur.
Deep Bacteria: Robust Deep Learning Data Augmentation Design for Limited Bacterial Colony Dataset
by Nour Eldeen M. Khalifa, Mohamed Hamed N. Taha, Aboul Ella Hassanien, Ahmed Abdelmonem Hemedan
Abstract: Bacterial colony classification is an important problem in microbiology. With the advances of computer-aided softwares, similar problems have been solved in a speedy and accurate manner during the last decade. In this paper, a deep neural network architecture will be presented to solve the bacterial colony classification problem. In addition, a training and testing strategy that relies on the strong use of data augmentation will be introduced. The used dataset was limited as it contains 660 images for 33 classes of a bacterial colony. Any neural network cant learn from this data directly and in case of learning the neural network will overfit. The adopted training and testing strategy lead to a significant improvement in the training and testing phases. It raised the dataset images to 6600 images for the training phase and 5940 images for verification phase. The proposed neural net-work with the adopted augmentation techniques achieved 98.22% in testing accuracy. A comparative result is presented, and the testing accuracy was compared with those of other related works. The proposed architecture outperformed the other related works in terms of its testing accuracy.
Keywords: Bacterial Colony Classification; Deep Convolutional Neural Networks; Data Augmentation.
A Neural-based Re-ranking Model for Chinese Named Entity Recognition
by Jing Guo, Yaxiong Han, Yongzhen Ke
Abstract: Chinese Named Entity Recognition (CNER) is different from English Named Entity Recognition (ENER). There is no specific delimiter in Chinese text to determine the words in a sentence. Besides, the combination of Chinese text has a strong arbitrariness. These special cases usually bring more errors to the Chinese NER (CNER). We propose a re-ranking model based on BILSTM network and without using any other auxiliary methods. Our approach uses N-best generalized label sequences that are produced by baseline model as input and feeds them into our re-ranking model for modeling the context within the generalized sequences. The optimal output sequence is obtained by comprehensively considering the result of baseline model and re-ranking model. Experimental results show that our model achieves better F1-score on Bakeoff-3 MSRA corpus than the best previous experimental results, which yields a 0.97% improvement on F1-score over our neural baseline model and a 0.22% improvement over the state-of-the-art CNER model.
Keywords: chinese named entity recognition; computational linguistics; text recognition; neural architecture; deep learning.
Image Stitching Algorithm Based on Fast Feature Extraction and Perceptual Hash
by Zhe Wei, Aomei Li, Wanli Jiang
Abstract: The classic feature point image stitching algorithm is time-consuming in feature extraction, and an image sparse matrix method is proposed to determine the feature points. This method first uses Laplace operator to extract the image gradient and set the threshold to obtain the sparse matrix of image segmentation, then using Features from Accelerated Segment Test(FAST) detection algorithm for feature point . Finally, the Speeded-Up Robust Features(SURF) descriptor is given to increase the stability of the feature points. This method solves the problem of time-consuming feature extraction, making the real-time image mosaic possible. In the process of image mosaic, this paper presents a method using Kalman filter to predict the overlap region, thus reducing the computational complexity. In addition, because there is always a mismatch between feature point matching, this paper proposes a method using perceptual hashing to refine the matching pair, which solves the error caused by mismatching for the next image registration and improves the registration accuracy. Experimental results show that the proposed algorithm improves the speed and precision of image mosaic.
Keywords: Image mosaic; sparse matrix; Laplace operator; FAST; Kalman filter; perception hash.
Onto-Agent-SSSN: An ontology model to facilitate reactive reasoning in multi-agent systems within a business intelligence network
by Yman Chemlal
Abstract: Business intelligence network integrates the notion of collective intelligence to allow groups of individual collaborate collectively to produce useful information deployed by an organization. This useful information has become an extremely rich and complex asset on which decisions and strategies are based. Different techniques have been proposed to help the collaboration of the actors to produce collective intelligence. However, the most sophisticated tools will not be able to overcome a system where the actors do not want to communicate or have not acquired the reflex to exchange. At the sight of these elements, we proposed a platform based on a multi agent system called Agent-SSSN whose agents have a semantic structure guided by the semantic Web technologies (semantic agents) to reason as human experts. This platform integrates an agent-oriented modeling approach of the collaborative tasks of the actors in a network (interpretation of the collected information, deduction of the new knowledge necessary for the decision making and the intelligent diffusion of this information). The design of an Onto-Agent-SSSN ontology will be an encouraging solution to organize the knowledge in Agent-SSSN and facilitate the reasoning ability of the agent. Onto-Agent-SSSN ontology is implemented in Prot
Keywords: Business intelligence; business intelligence network; collective intelligence; multi-agent intelligent system; logical rules; ontology.
Special Issue on: Artificial Intelligent Techniques Applied to the Study of Engineering Applications
Evaluation of Worker Quality in Crowdsourcing System on Hadoop Platform
by Kavitha C, Srividhya Lakshmi R, Anjana Devi J, Pradheeba U
Abstract: Crowdsourcing is a new emerging distributed computing and problem solving production model on the backdrop of internet. The datasize of crowdsources and tasks grows rapidly due to the rapid development of the crowdsourcing system. To evaluate the worker quality, based on the big data technology has become a more complex challenge. In this paper, we propose a general worker quality evaluation algorithm which can be applied to any critical tasks without wasting resources. Realizing the evaluation algorithm in the hadoop platform using MapReduce parallel programming is also involved. Efficiency and accuracy of the algorithm is effectively verified in the wide variety of many big data scenarios.
Keywords: crowdsourcing system; hadoop; mapreduce.
Design and Implementation of Finite State Machine Using Quantum-Dot Cellular Automata (QCA)
by Sungeetha .D, Keerthana G, Vijayakumar K
Abstract: Moores Law states that the number of transistors per square inch on integrated circuits has doubled approximately every two years, this is true for CMOS based VLSI circuit design. Quantum-Dot Cellular Automata (QCA) replaces CMOS based VLSI technology. The assembly of quantum dots replaces transistors which is said to be Quantum Dot Cellular Automata, an emerging nanotechnology in the field of quantum electronics. Such type of circuit can be used in many digital applications and has an advantage of reduced area utilization. Quantum mechanics and cellular automata are together said to be Quantum Dot Cellular Automata. QCA technology has advantages like small size and high speed. CMOS technology uses transistors to create a logic gates but in QCA technology, logic gates and wires are created by using QCA cells. The basic logic gates like AND, OR, inverter, majority gates are implemented. Many combinational and sequential circuits are designed by using these basic gates. This paper aims at the design of Finite State Machines and its use in Vending machine and Traffic Light Controller were discussed. The circuit was designed and the functionality of those was verified using QCADesigner tool.
Keywords: QCA; Finite state machine; Vending machine; Traffic Light Controller; QCADesigner 2.3.0.
IDENTIFICATION OF PERSON OR DATA USING MODIFIED SQUARE BLOCKWISE APPROACH
by Denslin Braja R, Dharun V. S.
Abstract: Nowadays authentication is important to identify the originality of a person or document. Visual Secret Sharing scheme is one of the best methods to provide authentication without any complex computations. In this paper, we proposed novel method called modified square blockwise approach to generate shares. Here the personage photo or any biometric data can be used as an authenticated image. This approach will generate two shares, one share is printed on identity card and the other one is stored on database. To verify the originality of a document or a person, first receive the identity card and scan or take a photo of the share. Now this share is compared with the stored one, if it is reveal the authenticated image then it is accepted, otherwise simply rejected. This approach is used to authenticate any confidential data such as medical document, bank details and administration details also. Using this approach, we can restrict the use of confidential documents without knowing the originality of the document.
Keywords: Authentication; Modified square blockwise approach; Visual Secret Sharing.
A Secured Cloud Storage Auditing with Empirical Outsourcing of Key Updates
by Vijayakumar K, Suchitra S, Swathi Shri P
Abstract: Cloud Computing is emerging and considered next generation architecture for computing. Cloud storage auditing is regarded as a prominent feature to validate the integrity of the data in public cloud. The key exposure resistance is a vital issue in various security applications. The current cloud storage auditing models require the client to update his secret keys in each time period which bring in local burdens to the client. The aim of the project is to make the key updates transpicuous to the client and present a new model for auditing cloud storage with verifiable outsourcing of key updates. In this model, the key updates are outsourced to a third party auditor (TPA) who reduces the local burden on the client. The third party auditor (TPA) is accountable for both cloud storage auditing and safe key updates. The TPA holds an encrypted version of the clients secret key. The client decrypts the secret key, generates authenticators for the file and uploads these files along with authenticators to the cloud. In our design, we employ the Multi Key Encryption Algorithm to achieve faster key updates, short key size and to proficiently encrypt the secret keys. In addition, the TPA will audit whether the files in cloud are stored correctly by a challenge-response protocol between it and the cloud at regular time. The proposed paradigm permits the client to authenticate the validity of the encrypted secret key produced by the third party auditor for uploading data to the cloud. These prominent features are considered to make the entire cloud storage auditing technique as transpicuous as possible for the client.
Keywords: Cloud Computing; Encryption; Decryption; Multi Key Encryption Algorithm.
Stimulated RR MAC Protocol for Power Efficient Wireless Sensor Networks
by Kirubakaran M.K., Sankarram N.
Abstract: Wireless Sensor Networks is a collection of sensor nodes scattered across a habitat of interest to collect crucial information from the habitat. The application of these networks is enormous, ranging from video surveillance, medical device monitoring, air traffic control, robot control, target monitoring, border protection, disaster assessments etc. Since the nodes in wireless sensor networks remain in the habitat for long duration, it is necessary they utilize the battery mounted onboard very efficiently. Research has grown along this area where in many protocols are being proposed in order to enable the nodes use the energy efficiently. This paper proposes one such protocol Stimulated RR MAC protocol and discusses its benefits and performance over other existing protocols.
Keywords: Channel allocation; Channel capacity Multipath channels; Multicast protocols; Wireless communication.
Students Performance Analysis System Using Cumulative Predictor Algorithm
by DafniRose J, VIJAYAKUMAR K, Sakthivel Srinivasan
Abstract: The recent trends in the IT industry indicates that it is moving towards automation to do mundane tasks and the expectations for students already equipped with good programming skills is on the rise. In parallel, there arehas been a rising number of students who find it difficult to attain the skills necessary in order to get the dream IT job they desire. The aim of this project is to bridge the gap between the employer and the future employee of the company by the use of SPAS at college level.Student Performance Analysis System (SPAS) is an online web application system which enables students to know prior hand if their level of skills for the placement is enough to get placed or not, given the necessary inputs.SPAS has an intelligent learning algorithm which utilizes a rich database, analyses the records of previous students traits and develops a model for further prediction.The performance evaluation of students by SPAS is by the cumulative predictor algorithm involving generation of several random forest trees on the available data. SPAS learns and creates its model reaching higher accuracy with increasing data availability.
Keywords: Educational Data mining; decision tree (J48); Naïve Bayes’ classifier; JRip algorithm; bagging method; standard deviation; infogain; entropy gain.
A Hybrid Algorithm for efficient Task Scheduling in Cloud Computing Environment
by Roshni .T, Uma Maheswari, Bijolin Edwin
Abstract: Cloud is a boon to the generation which provides services that can reduce the overhead in maintenance and computational complexities.Scheduling the users job in the cloud resources plays an important role for the better performance. Task scheduling is an NP-Hard problem, since it may have more than one solution to fit in. In this paper a hybrid algorithm is proposed by the amalgamation of Artificial Bee Colony Algorithm and Particle Swarm Optimization named as ABPS algorithm. The proposed ABPS algorithm optimizes the task scheduling on the cloud environment by providing minimized makespan, cost, and maximized resource utilization and to balance the load. The experiments were simulated using cloudSim tool and the ABPS algorithm results outperform the original ABC and PSO algorithm. The comparative study proves the performance of the proposed ABPS algorithm over the original ABC and PSO algorithm.
Keywords: Cloud Computing; Task Scheduling; ABC algorithm; PSO algorithm; Makespan; Cost; Load Balancing; Resource Utilization.
A Survey on Contrastive Opinion Summarization
by Lavanya SK
Abstract: Contrastive OpinionSummarization (COS) is jointly generating summaries for two entities in order to highlight their differences based on the features.COS comprises of feature extraction, Sentiment prediction and summarization.Recently, the research focus in COS has been in using semantics associated with words and multi-word expressions to shift from syntactic to semantic level. This survey paper covers different methods used forfeature extraction, various similarity measures and different types of summarization. In addition to these, various datasets and performance measures are also addressed. Finally, future research directions are also suggested.
Keywords: Opinion mining; Feature based opinion summarization; Feature extraction; Sentiment prediction,Contrastive Opinion Summarization (COS).
ANALYSIS OF HEURISTIC BASED MULTILEVEL THRESHOLDING METHODS FOR IMAGE SEGMENTATION USING R PROGRAMMING
by Suresh .K, Sakthi .U
Abstract: The conventional way in analyzing image segmentation algorithms manually is difficult since it requires a lot of human effort in keeping all data for analysis. Various heuristic algorithms are bundled with Otsus and Kapurs objective function in finding optimal fitness and quality segmentation. In this work Otsus and Kapurs objective function are bundled with heuristics such as Harmony Search Optimization (HSO) and Electro Magnetic Optimization (EMO) to compare the solution accuracy of segmented images In order to statistically analyze such algorithms, an automated tool is developed which takes an input image of any image category under consideration and extracts the segmentedfitness values and quality parameters of the image. The extracted values are stored in a central database server constrained with image type, image category, methodology and heuristic used, no of thresholds and quality parameters. The central repository information is fed into data mining and data analytic tools to statistically rank the segmentation algorithms.
Keywords: Otsu &Kapur Objective function; Electro-magnetic Optimization; Harmony Search Optimization; Rank test.
Aggregated Clustering for Grouping of Users based on Web Page Navigation Behavior Aggregated Clustering for User Grouping
by GeethaRamani .R, Revathy .P, Lakshmi B
Abstract: In this epoch, a significant amount of patterns are retrieved using data mining techniques. Application of data mining techniques to the World Wide Web is referred as Web Mining. Clustering is one of the data mining technique that plays an vital role in the field of Web mining. This paper works on the server logs from the MSNBC dataset for the month of September 1999. Users with the average access length of 6 are used for analysis. This research aims to cluster the user based on their navigation behavior. An iterative aggregated clustering is proposed, in which various clustering algorithms such as EM clustering, Farthest First, K-Means Clustering, Density based cluster, Filtered cluster are applied on the dataset. The resultant clusters from various algorithms are aggregated correspondingly and the frequency of instances in each cluster is determined. If frequency of a instance in a cluster is greater than or equal to two-third majority, then the instance is grouped in that cluster. The work revealed that the system guaranteed to cluster 91% of users in the first iteration under 17 clusters for each page category and 99% of users are clustered in the subsequent iterations in another 17 clusters and rest of the users are grouped as one cluster, resulting in 35 hard clusters. The proposed framework is believed to serve in clustering user groups there by enabling suitable customized web environment.
Keywords: Clustering algorithm; Data mining; MSNBC; Web usage mining; Hard Clusters.
Unified Dynamic Texture Segmentation System based on Local and Global Spatiotemporal Techniques
by Shilpa Paygude, Vibha Vyas
Abstract: Dynamic Texture (DT) is temporal extension of static texture. There are two broad types of dynamic textures: Natural and Manmade. The examples of dynamic textures are waving tree, sea water, fountain, traffic, moving crowd etc. Dynamic Texture Segmentation is a technique used to separate the moving objects from stationary content in the video. Majority of the techniques give good results either for natural or for manmade dynamic textures. The proposed approach for DT Segmentation is combination of Local Spatiotemporal Technique i.e. Local Binary Pattern-Weber Local Descriptor and Global Spatiotemporal Technique i.e. Contourlet Transform. The local spatiotemporal technique considers the appearance and motion of the object for segmentation. The technique is computationally less complex than optical flow and gives good results for manmade dynamic textures. The Global spatiotemporal technique is based on Laplacian Pyramid and Directional Filters. It gives good results for natural dynamic textures. The proposed technique discussed in this paper is unified approach for any type of Dynamic Texture. The regions of images commonly segmented by both the techniques are considered as the final segmented output. The proposed system works equally well on any kind of Dynamic Texture.
Keywords: Dynamic Texture; Segmentation; Optical Flow; LBP-WLD-OF; Contourlet Transform; Unified Approach.
Enhanced H-ABE-R and T-ABKS for Cloud Computing Environment
by Rachel .N, C. R. Rene Robin
Abstract: For a data owner (patient) wants to share his health data, data which is extremely sensitive in nature and when he wishes to maintain privacy it is common for data owners to encrypt the data. Sharing encrypted data leads to difficulties in key management (traditional PKI infrastructure).Boneh et al first put forth identity based encryption that helped alleviate those problems. However, Identity based encryption did not provide solutions for revocation users in a multi-user setting. The problem was tackled by Li et al. In the first half of our work, we propose an novel attribute based encryption, a further extension of Identity based encryption that allows roles or policies to be defined over a set of attributes and propose a Hybrid Attribute based Encryption with Revocation (H-ABE-R) which is more suitable for multi-user hospital settings as it allows us to give permissions to access data for short periods of time. We then propose a Time & Attribute based keyword search (T-ABKS). We use Dynamic Bloom filters (DBF) for to represent dynamic datasets(datasets which reflect details like blood sugar, blood pressure tend to change every day) and discuss a framework for searching over masked bloom filters. Further by masking of random numbers the DBF we provide security(S-DBF) and prevent from correlation attacks. We check our algorithm and and prove our scheme comparably efficient.
Keywords: Verifiable Computation; keyword search; Attribute Based Encryption; Revocation.
Special Issue on: ICCD-2017 Internet of Things, Big Data and Machine Learning
Evaluation Research on Green Degree of Equipment Manufacturing Industry Based on Improved Particle Swarm Optimization Algorithm
by Rong Wang, Ying Wang
Abstract: In order to improve the sustainable development of equipment manufacturing industry, the improved particle swarm algorithm is applied in evaluating green degree of equipment manufacturing industry. Firstly, the green degree evaluation system of equipment manufacturing industry is constructed, and evaluation index system is established. Secondly, the basic theory of particle swarm algorithm and the improved particle swarm algorithm are studied basing on analysis of disadvantages of traditional particle swarm algorithm. Thirdly, the analysis procedure of improved particle swarm algorithm is designed. Finally, equipment manufacturing industry in a province is used as a researching object, the green degree evaluation of equipment manufacturing industry in this province is carried out, and results show that this algorithm can improve evaluation level of green degree of equipment manufacturing industry.
Keywords: green degree; equipment manufacturing industry; improved particle swarm algorithm.
Network Security Situation Detection of Internet of Things for Smart City Based on Fuzzy Neural Network
by Qing Liu, Ming ZENG
Abstract: in order to ensure the safety of Internet of things for smart city and ensure normal operation of smart city, the network security situation of Internet of things should be monitored correctly for a long time, therefore the fuzzy neutral network with wavelet package and chaos particle swarm algorithm is applied it. Firstly, the basic theory of network security situation of Internet of things for smart city is analyzed, the corresponding mathematical is constructed, and the security situation awareness framework of Internet of things is designed. Secondly, the basic theory of fuzzy neutral network is studied, and the structure of the fuzzy neutral network is designed. Thirdly, the processing method of network security situation data based on wavelet package is constructed. And then training procedure of fuzzy neutral network based on chaos particle swarm algorithm is established, the algorithm procedure is designed. Finally, the simulation analysis is carried out using a smart city as example, and the network security situation of Internet of things for it is monitored correctly, then the network safety can be ensured.
Keywords: fuzzy neutral network; smart city; Internet of things; network security.
Special Issue on: ICICT2018 Advances in Intelligent Information Communication Technologies
Multi-Criteria Clustering-based Recommendation using Mahalanobis distance
by Mohammed Wasid, Rashid Ali
Abstract: There have been significant advances made in the research of recommender systems over the past decades and have been implemented in both industry and academia. Recently, multi-criteria ratings are being incorporated into traditional recommender systems to further improve their quality, especially to handle the data sparsity and cold start issues. However, incorporation of multi-criteria ratings have improved the performance of the recommendation, but at the same time, multidimensionality issue is also arises. This paper presents a clustering based recommendation approach which is used for dealing with the multi-dimensionality issue in multi-criteria recommender systems. Here, we cluster the users based on their individual criteria ratings using K-means cluster-ing and the intra-cluster similarity is computed using Mahalanobis distance measure for neighborhood set gen-eration. This improves the recommendations quality and predictive accuracy of both traditional and clustering-based collaborative recommendations. The Yahoo! Movies dataset was used for testing the approach and the experiment conducted shows promising results.
Keywords: Recommender systems; RS; Collaborative filtering; CF; Mahalanobis distance; MD; K-Means clustering; Multi-criteria.
Fast Algorithm of Image Enhancement based on Multi-Scale Retinex
by Alexander Zotin
Abstract: In this paper, a fast image enhancement algorithm based on Multi-Scale Retinex in HSV color model is presented. The proposed algorithm produces the result similar to the one which uses a nonlinear processing in the HSV color model, but with less computational cost. It uses linear dependencies of RGB colors from the V-component of HSV model. Additionally, to speed up the images processing and enhance the local contrast is suggested to perform Multi-Scale Retinex (MSR) computation only in the low-frequency area obtained by the wavelet transform. Experimental research was performed on more than 100 color images having non-uniform brightness. Different algorithms based on Retinex technology were implemented and their performance was compared. The proposed way of output image color formation allows to reduce processing time by 30-75%, depending on the image size. The experimental data show that the usage of the wavelet transform in proposed MSR algorithm additionally leads to 2-2.8 times increase in processing speed.
Keywords: Color image enhancement; Retinex; MSR; Multi-Scale Retinex; Color space; HSV; Wavelet transform;.
Exchanging Deep knowledge for fault diagnosis using ontologies
by Xilang Tang, Mingqing Xiao, Bin Hu, Dongqing Pan
Abstract: To improve the development efficiency of automated diagnosis equipment (ADE) and ensure the generality of ADE software, this paper proposes a novel method to exchange deep knowledge of systems under diagnosis (SUD) using ontologies. A general framework of knowledge base combining test information model and diagnosis information model is proposed. The diagnosis information model is decomposed into structure model and function model. The structure model describes the connectivity of adjacent components as well as the structural hierarchy, and the function model describes behavior of modules by mapping input signals into output signals. Moreover, the method to locate the fault based on the proposed knowledge base is introduced. Finally, a case study for guiding system of passive-radar guidance missile is carried out to illustrate our proposed method. The practice shows that our method can achieve the object well
Keywords: fault diagnosis; test; knowledge; ontology; reasoning.
Multistage approach for automatic spleen segmentation in MRI sequences
by Antonia Mihaylova, Veska Georgieva, Plamen Petrov
Abstract: Most of the known methods of segmentation of the abdominal organs are not automated for the whole series of images or are semi-automatic and require additional intervention by the user. This is typical for cases where the difference in intensity of the gray level between the subject and the background is small. A typical example of this is the spleen and adjacent tissue in unconstrained MR images. This paper presents a multistage approach for spleen segmentation from MRI-sequences. It is based on segmentation methods such as active contours without edges and k-mean clustering. The proposed approach consists of some basic stages. The first stage is pre-processing, based on image enhancement and morphological operation. Two atlas models are created, which are used in the initial image to define the initial contour at which the segmentation begins. The initial image is semi-automatic segmented using the created atlas models. The sequence is then automatic segmented, dividing it in two parts (before and after the initial middle image) and using the segmentation of the previous image. The proposed approach allows extracting the spleen in the different depth images, which has a variable form and unstable position. The conducted experiments are showing the robustness of the proposed approach. The obtained results demonstrate the effectiveness of the approach for application in screening diagnostics.
Keywords: Segmentation of Spleen; Segmentation of MRI sequences; Automatic Segmentation.
Classification of Radar Non-Homogenous Clutter Based on Statistical Features Using Neural Network
by Thamir Saeed, Ghufran Hatem, Jafar Abdul Sadah
Abstract: This paper presents a robust clutter classifier based on the neural network to assist the radar receiver by choosing optimal constant false alarm rate. Where this classifier has been trained for sixteen class, four radar return distribution with different situations. The return radar signal distributions are Rayleigh, Weibull, lognormal and K- distribution, while the situations are, Signal, Multi-, Closed Multi-target, and clutter edge. Multi-layer perceptron with back-propagation as a neural network with seven features, Mean, Variance, Mode, Kurtosis, Skewness, Median, and Entropy, have been used to classify the return signal. A Least mean square error is used to evaluate the classifier performance. The simulation is evaluated for the Signal to clutter ration from +35dB to -35 dB, with 5-20 neurons of the hidden layer, and 60-360 samples. By performing, the Optimization has been gained by using 240 samples and 20 neurons then lead to 98.1 % return signal classification
Keywords: Clutter Classifier; CFAR; Radar; and Non-homogenous clutter; statistical Features.
Development of a sit-to-stand assistance chair for elderly people
by Ari Aharari, Won-Seok Yang
Abstract: According to the survey on the actual situation of elderly persons at home or nursing home care, the first item after concerning about disease under treatment is Weak legs and difficulties to stand from the chair. Muscle strength further decreases with aging and make feeling burden when standing from chair. Also, people who are suffering from secondary symptoms such as bedsores and keep sitting in a chair for a long time are on the rise. The most burdensome for elderly persons when trying to stand up from the chair is to bear the weight themselves. In this paper, we introduce Rakutateru which is specially designed to support elderly persons to easily stand up from the chair and keep people to more active and independent. We also evaluate the validity of an assist unit which is contained inside the lower part of the Rakutateru surface.
Keywords: Assist chair; Elderly support chair; Lifting unit.
Onboard Reasoning and Other Applications of the Logic-Based Approach to the Moving Objects Intelligent Control
by Andrey Tyugashev
Abstract: This article provides the theoretical background and practical case studies of the application of reasoning and other logic-based approaches to the moving objects control. Modern moving objects, both manned and unmanned, utilize computers as their onboard brain. Since planes, spacecraft, cars, trucks and trains must demonstrate flexible and safe behavior in various situations, it seems prospective to use intelligent control means instead of rigid control logic dispersed in a program source code. This article is concerned with the possible implementation of onboard intelligence. In contrast to the popular use of neural networks, the logic-based approach is based on clear and exact control rules with strict responsibility. Thus, formal specification and verification methods can be utilized. The article describes the Real-Time Control Algorithm Logic (RTCAL) for the above-mentioned purposes. We also present case studies of reasoning at the design and operation stages for providing the fault tolerant control of a spacecraft.
Keywords: Moving objects control; logic; intelligent control; reasoning; Real-Time Control Algorithm; flight control software.
Special Issue on: ICEST'18 Intelligent Sensor Data Processing, Mobile Telecommunications and Air Traffic Control
Application Level Extension of Bandwidth Management in Radio Access Network
by Evelina Pencheva, Ivaylo Atanasov
Abstract: Multi-access Edge Computing (MEC) provides processing and storage capabilities of the cloud into the radio access network. In this paper, we study the deployment of bandwidth management service in MEC environment. The bandwidth management service procedures are mapped onto functionality of the control protocol between radio access network and core network. An extension of the bandwidth management service is proposed that enables detecting of packets generated of specific applications and applying the appropriate enforcement actions. The proposed extension is described by typical use cases, information flows, required information, data model, as well as respective application programming interfaces. Models representing the status of bandwidth allocation as seen by the mobile edge application and network are proposed, formally described and verified. Formal model verification enables mathematical demonstration that the proposed extension is consistently implementable.
Keywords: Quality of service control; Bandwidth management; Application detection and control; Radio access network; Multi-access Edge Computing; Application Programming Interfaces; Data model; Finite state machines.
Flight Safety Sensor and Auto-Landing System of Unmanned Aerial System
by Krume Andreev, Georgi Stanchev
Abstract: Over the past decade, there has been a rapid development of Unmanned Aerial Systems (UAS). The trend and current developments lead to an increase in the use of UAS. The operations of UAS and their use significantly increase every day. This article provides solutions and options for introducing a flight safety sensor system and auto-landing system for UAS. The reason is to ensure effective completion of their mission without the involvement of a qualified operator (pilot) in the control station. The problems and characteristics of these systems and the algorithms through which they successfully perform their tasks are analyzed in the article. In the article has been proposed an architectural realization of a flight safety sensor system and an auto- landing system for UAS.
Keywords: Flight Safety; Sensor System; Technical Condition; Auto-Landing System; Unmanned Aerial System; Conical Scanning; Pseudo-Conical Scanning.
Performance of VWM algorithm in the presence of impulse noise and resizing
by Bojan Prlincevic, Zoran Milivojevic, Stefan Panic
Abstract: The first part of this paper describes VWM (Visible Watermarking) algorithm for inserting and removing visible watermark in the image. The second part of this paper describes an experiment in which the image is watermarked with the VWM algorithm,impulse noise is added, and the image quality is improved with the MDB algorithm for filtering. Watermark is removed from noised andfiltered image. Afterwards, an experiment is described in which resizing of the noised watermarked image is performed. Watermark is removed from this image. Finally, a comparative analysis of the results is performed in order to evaluate the efficiency of the applied algorithms. The comparison was performed on the basis of MSE and Similarity. The obtained results are analysed in detail and presented in a tabular and graphical manner.
Keywords: Visible watermark; Impulsive noise; Filtering; Resizing;.
Design and optimization of bio-inspired robotic stochastic search strategy
by Farhad Maroofkhani
Abstract: An autonomous robots search strategy is the set of rules that it employs while looking for targets in its environment. Biological systems (e.g., foraging animals) provide useful inspirations for designing optimal stochastic search algorithms for autonomous robots. Due to the complexity of interaction between the robot and its environment, optimization must performed in high-dimensional parameter space. We analyze the dependence of search efficiency on environmental parameters and robot characteristics using Response Surface Methodology (RSM), a technique originally developed for experimental design. In this study, the efficiency of a strategy focuses on L
Keywords: Levy walk; Autonomous robots; Swarm robot; Biomimetic; Individual motion; Design of experiments.
Influence of optimal pair-wise SUS algorithm on MU-MIMO-OFDM system performance
by Aleksandra Panajotovic, Nikola Sekulovic, Daniela Milovic
Abstract: In this paper we proposed a new user scheduling algorithm, named as optimal pair-wise semi-orthogonal user selection (SUS), for multiuser multiple-input multiple-output orthogonal frequency division multiplexing (MU-MIMO-OFDM) system. Multiuser interferences are canceled applying zero-forcing beamforming (ZFBF) technique with presumption that channel state is perfectly known at transmit side. Simulated throughput and error results demonstrate advantage gain achieved in system performance realized through applying the proposed scheduling algorithm.
Keywords: FLA; IEEE 802.11ac; MU-MIMO-OFDM; User Scheduling Algorithm; ZFBF.
The effect of background and outlier subtraction on the structural entropy of two-dimensional measured data
by Szilvia Nagy, Brigitta Sziova, Levente Solecki
Abstract: For colonoscopy images the main information is in the fine structure of the surface of the bowel or colorectal polyps, similarly to the case of combustion engine cylinder surface scans, where the grooving and wear can be detected from the fine pattern superposed to a cylinder curvature.
In both cases appear outliers, colonoscopy images have many reflections, whereas the roughness scanners detect small dust particles as well as the micron scale vibrations from the environment.
The method presented in this paper takes care of both the problems using histogram stretching together with a special type of filtering. Also, masks are introduced in order to control the effect of the operators.
The effects of the processing steps on the structural entropy of the image is also studied, as structural entropies are used in characterization of the images. By removing the background makes the structural entropies much smaller, and by suppressing the outliers the structural entropies increase.
Keywords: Image preprocessing; Rényi entropy; structural entropy; colonoscopy; microgeometrical surface.
Special Issue on: Modeling as a Service for Designing and Analyzing QoS-Oriented Information, Data and Knowledge Systems
Machine Learning Methods Against False Data Injection In Smart Grid
by MOHAMED HAMLICH
Abstract: The false data injection in the power grid is a major risk for a good and safety functioning of the smart grid. The False data detection with conventional methods are incapable to detect some false measurements, to remedy this, we have opted to use machine learning which we used Five classifiers to conceive an effective detection (k-nearest neighbor algorithm "KNN", Random trees, Random forest decision trees, multi-layer perceptron and vector support machine). Our analyze are validated by experiments on a physical bus feeding system performed on PSS / in which we have developed a data set for real measurement. Afterward we worked with Matlab software to construct false measurements according to the Jacobean matrix of the state estimation. We tested the collected data with different classification algorithms, which gives good and satisfactory results.
Keywords: smart grid;
false data injection;