International Journal of Grid and Utility Computing (11 papers in press)

Regular Issues

• Recommendation system based on space-time user similarity
  by Wei Luo, Zhihao Peng, Ansheng Deng 
  Abstract: With the advent of 5G, the way people get information and the means of information transmission have become more and more important. As the main platform of information transmission, social media not only brings convenience to people's lives, but also generates huge amounts of redundant information because of the speed of information updating. In order to meet the personalised needs of users and enable users to find interesting information in a large volume of data, recommendation systems emerged as the times require. Recommendation systems, as an important tool to help users to filter internet information, play an extremely important role in both academia and industry. The traditional recommendation system assumes that all users are independent. In this paper, in order to improve the prediction accuracy, a recommendation system based on space-time user similarity is proposed. The experimental results on Sina Weibo dataset show that, compared with the traditional collaborative filtering recommendation system based on user similarity, the proposed method has better performance in precision, recall and F-measure evaluation value.
  Keywords: time-based user similarity; space-based user similarity; recommendation system; user preference; collaborative filtering.
  
  
• Joint end-to-end recognition deep network and data augmentation for industrial mould number recognition  
  by RuiMing Li, ChaoJun Dong, JiaCong Chen, YiKui Zhai 
  Abstract: With the booming manufacturing industry, the significance of mould management is increasing. At present, manual management is gradually eliminated owing to need for a large amount of labour, while the effect of a radiofrequency identification (RFID) system is not ideal, which is limited by the characteristics of the metal, such as rust and erosion. Fortunately, the rise of convolutional neural networks (CNNs) brings down to the solution of mould management from the perspective of images that management by identifying the digital number on the mould. Yet there is no trace of a public database for mould recognition, and there is no special recognition method in this field. To address this problem, this paper first presents a novel data set aiming to support the CNN training. The images in the database are collected in the real scene and finely manually labelled, which can train an effective recognition model and generalise to the actual scenario. Besides, we combined the mainstream text spotter and the data augmentation specifically designed for the real world, and found that it has a considerable effect on mould recognition.
  Keywords: mould recognition database; text spotter; mould recognition; data augmentation.
  
  
• University ranking approach with bibliometrics and augmented social perception data  
  by Kittayaporn Chantaranimi, Rattasit Sukhahuta, Juggapong Natwichai 
  Abstract: Typically, universities aim to achieve a high position in ranking systems for their reputation. However, self-evaluating rankings could be costly because the indicators are not only from bibliometrics, but also the results of over a thousand surveys. In this paper, we propose a novel approach to estimate university rankings based on traditional data, i.e., bibliometrics, and non-traditional data, i.e., Altmetric Attention Score, and Sustainable Development Goals indicators. Our approach estimates subject-areas rankings in Arts & Humanities, Engineering & Technology, Life Sciences & Medicine, Natural Sciences, and Social Sciences & Management. Then, by using Spearman rank-order correlation and overlapping rate, our results are evaluated by comparing with the QS subject ranking. From the result, our approach, particularly the top-10 ranking, performed estimating effectively and then could assist stakeholders in estimating the university's position when the survey is not available.
  Keywords: university ranking; rank similarity; bibliometrics; augmented social perception data; sustainable development goals; Altmetrics.
  
  
• Assessment of a cuckoo search-based intelligent system for mesh routers placement optimisation in WMNs considering various distributions of mesh clients  
  by Shinji Sakamoto 
  Abstract: Wireless Mesh Networks (WMNs) have many advantages. However, they have several issues related to wireless communication. An effective approach to deal with these problems is the optimization of mesh routers placement in WMNs, but this is an NP-hard problem. Thus, heuristic and intelligent algorithms are needed. In the previous work, we developed an intelligent simulation system based on Cuckoo Search (CS) (WMN-CS), which is a meta-heuristic algorithm. In this work, we evaluate the WMN-CS performance for various distributions of mesh clients: Uniform, Normal, Exponential, Weibull and Chi-square distributions. The simulation results show that for Normal distribution the WMN-CS system can find suitable locations of mesh routers for about 30 phases. The Uniform distribution has the lowest performance compared to the other distributions. Also, Exponential andWeibull distributions converge slower than Chi-square distribution. However, Exponential distribution converges faster than Weibull distribution.
  Keywords: wireless mesh networks; node placement problem; cuckoo search; client distributions.
  DOI: 10.1504/IJGUC.2024.10068275
   
  
• Analysis of cybersecurity attacks and solution approaches  
  by Ali Yılmaz, Resul Das 
  Abstract: Numerous cybersecurity vulnerabilities have emerged as a result of the quickly changing digital environment, posing serious risks to people, corporations, and governments alike The article "An analysis cybersecurity attacks and solution approaches" delves into the subtleties of this contemporary conflict zone and offers a thorough analysis of the many assault vectors, techniques, and their far-reaching effects This article examines the full spectrum of online hazards, from well-known enemies such as malware, phishing, and denial-of-service attacks to more sophisticated and sneaky threats such as advanced persistent threats (APT) It explores the methods and strategies employed by cybercriminals, offering light on their always-changing approaches This article also sheds light on preventive measures and problem-solving techniques created to combat these risks Provides an overview of a wide range of cybersecurity techniques and tools, including artificial intelligence, intrusion detection systems, encryption, and multifactor authentication Individuals and organizations can protect their digital domains from the onslaught of cyberattacks by being aware of the threats and having the necessary tools and knowledge. With a clearer grasp of the cyberthreat landscape and the various solutions to reduce these risks, readers should be better equipped to use the internet safely as a result of this thorough analysis.
  Keywords: information security; cybersecurity; malware; network security; attacks; cyberthreats.
  DOI: 10.1504/IJGUC.2024.10068492
   
  
• A two-stage intrusion detection framework in IoT using random forest for binary and multi-class classification  
  by Arash Salehpour, Pejman Hosseinioun, MohammadAli Balafar 
  Abstract: The proliferation of IoT devices due to cyber threats, which have become increasingly sophisticated, requires a strong security framework. This paper proposed a new framework for Intrusion Detection System-IoT-IDs using a Random Forest classifier to first classify the attack into binary features and prepare a new dataset that would enable multiclass classification. It achieved an overall accuracy of 0.98 on the comprehensive UNSW-NB15 dataset, with very good performance in detecting 'Generic' attacks, having almost perfect precision, recall, and F1-score. it also presented cases of 'Analysis' and 'Backdoor' types of attacks, where further improvements should be done. All these models have been analyzed to find the pros and cons in IoT settings. The Random Forests, XGBoost and MLP. Further studies based on the research could be done on multiplying models with improved features, intrusion detection in real time, and more strong AI techniques. This paper focuses on addressing challenges with imbalanced classes and scalability concerns using data privacy preservation methods for improving the performance of IDS.
  Keywords: IDS; intrusion detection system; IoT; internet of things; ensemble learning; UNSWNB15; cybersecurity; hybrid models.
  DOI: 10.1504/IJGUC.2024.10071203
   
  
• A fuzzy-based system for decision of driver mental status and its performance evaluation  
  by Yi Liu 
  Abstract: For safe driving and decreasing the risk of tra_c accidents, it is important to control the driver mental status. In this paper, it is implemented an intelligent system based on Fuzzy Logic (FL) for deciding Driver Mental Status (DMS). In order to investigate the e_ects of the considered parameters we implement two models: DMS Model 1 (DMSM1) and DMS Model 2 (DMSM2). The input parameters of DMSM1 include Driver Anxiety Level (DAL), Tra_c Situation (TS), Driving Operating Time (DOT), while for DMSM2 we add a new parameter called Drive Distress Situation (DDS). For both models, the output parameter is DMS. We compared the simulation results of DMSM1 and DMSM2. The evaluation results show that DMSM2 is more complex because the rule base is bigger than DMSM1, but it has a better decision of DMS value.
  Keywords: safe driving; driver mental status; fuzzy logic; intelligent algorithms; VANETs.
  DOI: 10.1504/IJGUC.2025.10074407
   
  
• Neural network data analysis and mathematical modelling for Wordle games  
  by Weijun Chen, Jiangtao Jin, Yaxin Lei, Jian Tao 
  Abstract: Wordle is a popular daily puzzle in the New York Times. The Predicting Wordle Results problem in the 2023 Mathematical Contest in Modelling (MCM) focused on developing a model to estimate the reported headcount in the difficult mode. The model considered three attributes: word frequency, letter repetition and letter frequency. The analysis showed that these attributes influenced the reported headcount. Correlation analysis revealed a strong relationship between the number of participants and word frequency and letter repetition. A neural network time series model was developed, using word frequency and letter frequency as inputs to predict the reported results. The model achieved a high accuracy with an R
  Keywords: neural network; time series; information theory; entropy topsis; distance discriminant method; data analysis.
  DOI: 10.1504/IJGUC.2024.10074408
   
  

Special Issue on: AMLDA 2022 Applied Machine Learning and Data Analytics Applications, Challenges, and Future Directions

• Fuzzy forests for feature selection in high-dimensional survey data: an application to the 2020 US Presidential Election  
  by Sreemanti Dey, R. Michael Alvarez 
  Abstract: An increasingly common methodological issue in the field of social science is high-dimensional and highly correlated datasets that are unamenable to the traditional deductive framework of study. Analysis of candidate choice in the 2020 Presidential Election is one area in which this issue presents itself: in order to test the many theories explaining the outcome of the election, it is necessary to use data such as the 2020 Cooperative Election Study Common Content, with hundreds of highly correlated features. We present the fuzzy forests algorithm, a variant of the popular random forests ensemble method, as an efficient way to reduce the feature space in such cases with minimal bias, while also maintaining predictive performance on par with common algorithms such as random forests and logit. Using fuzzy forests, we isolate the top correlates of candidate choice and find that partisan polarisation was the strongest factor driving the 2020 Presidential Election.
  Keywords: fuzzy forests; machine learning; ensemble methods; dimensionality reduction; American elections; candidate choice; correlation; partisanship; issue voting; Trump; Biden.
  
  
• An efficient intrusion detection system using unsupervised learning AutoEncoder  
  by N.D. Patel, B.M. Mehtre, Rajeev Wankar 
  Abstract: As attacks on the network environment are rapidly becoming more sophisticated and intelligent in recent years, the limitations of the existing signature-based intrusion detection system are becoming more evident. For new attacks such as Advanced Persistent Threat (APT), the signature pattern has a problem of poor generalisation performance. Research on intrusion detection systems based on machine learning is being actively conducted to solve this problem. However, the attack sample is collected less than the normal sample in the actual network environment, so it suffers a class imbalance problem. When a supervised learning-based anomaly detection model is trained with these data, the results are biased toward normal samples. In this paper, AutoEncoder (AE) is used to perform single-class anomaly detection to solve this imbalance problem. The experimental evaluation was conducted using the CIC-IDS2017 dataset, and the performance of the proposed method was compared with supervised models to evaluate the performance
  Keywords: intrusion detection system; advanced persistent threat; CICIDS2017; AutoEncoder; machine learning; data analytics.
  
  

Special Issue on: Cloud and Fog Computing for Corporate Entrepreneurship in the Digital Era

• Study on the economic consequences of enterprise financial sharing model  
  by Yu Yang, Zecheng Yin 
  Abstract: Using enterprise system ideas to examine the business process requirements of firms, the Financial Enterprise Model (FEM) is a demanding program. This major integrates finance, accounting, and other critical business processes. Conventional financial face difficulties due to low economic inclusion, restricted access to capital, lack of data, poor R&D expenditures, underdeveloped distribution channels, and so on. This paper mentions making, consuming, and redistributing goods through collaborative platform networks. These three instances highlight how ICTs (Information and Communication Technologies) can be exploited as a new source of company innovation. The sharing economy model can help social companies solve their market problems since social value can be embedded into their sharing economy cycles. As part of the ICT-based sharing economy, new business models for social entrepreneurship can be developed by employing creative and proactive platforms. Unlike most public organizations, double-bottom-line organizations can create social and economic advantages. There are implications for developing and propagating societal values based on these findings.
  Keywords: finance; economy; enterprise; ICT; social advantage.