Educational Data Mining (EDM) is a rich research field in computer science. Tools and techniques in EDM are useful to predict student performance which gives practitioners useful insights to develop appropriate intervention strategies to improve pass rates and increase retention. The performance of the state-of-the-art machine learning classifiers is very much dependent on the task at hand. Investigating support vector machines has been used extensively in classification problems; however, the extant of literature shows a gap in the application of linear support vector machines as a predictor of student performance. The aim of this study was to compare the performance of linear support vector machines with the performance of the state-of-the-art classical machine learning algorithms in order to determine the algorithm that would improve prediction of student performance. In this quantitative study, an experimental research design was used. Experiments were set up using feature selection on a publicly available dataset of 1000 alpha-numeric student records. Linear support vector machines benchmarked with ten categorical machine learning algorithms showed superior performance in predicting student performance. The results of this research showed that features like race, gender, and lunch influence performance in mathematics whilst access to lunch was the primary factor which influences reading and writing performance.

Ubiquitous learning (u-learning) refers to anytime and anywhere learning. U-learning has progressed to be considered a conventional teaching and learning approach in schools and is adopted to continue with the school curriculum when learners cannot attend schools for face-to-face lessons. Computer Science, namely the field of Artificial Intelligence (AI) presents tools and techniques to support the growth of u-learning and provide recommendations and insights to academic practitioners and AI researchers. Aim: The aim of this study was to conduct a meta-analysis of Artificial Intelligence works in ubiquitous learning environments and technologies to present state from the plethora of research. Method: The mining of related articles was devised according to the technique of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). The complement of included research articles was sourced from the broadly used databases, namely, Science Direct, Springer Link, Semantic Scholar, Academia, and IEEE. Results: A total of 16 scientific research publications were shortlisted for this study from 330 articles identified through database searching. Using random-effects model, the estimated pooled estimate of artificial intelligence works in ubiquitous learning environments and technologies reported was 10% (95% CI: 3%, 22%; I 2 = 99.46%, P = 0.00) which indicates the presence of considerable heterogeneity. Conclusion: It can be concluded based on the experimental results from the sub group analysis that machine learning studies [18% (95% CI: 11%, 25%), I 2 = 99.83%] was considerably more heterogeneous (I 2 = 99.83%) than intelligent decision support systems, intelligent systems and educational data mining. However, this does not mean that intelligent decision support systems, intelligent systems and educational data mining is not efficient.

In ordinary credit card datasets, there are far fewer fraudulent transactions than ordinary transactions. In dealing with the credit card imbalance problem, the ideal solution must have low bias and low variance. The paper aims to provide an in-depth experimental investigation of the effect of using a hybrid data-point approach to resolve the class misclassification problem in imbalanced credit card datasets. The goal of the research was to use a novel technique to manage unbalanced datasets to improve the effectiveness of machine learning algorithms in detecting fraud or anomalous patterns in huge volumes of financial transaction records where the class distribution was imbalanced. The paper proposed using random forest and a hybrid data-point approach combining feature selection with Near Miss-based undersampling technique. We assessed the proposed method on two imbalanced credit card datasets, namely, the European Credit Card dataset and the UCI Credit Card dataset. The experimental results were reported using performance matrices. We compared the classification results of logistic regression, support vector machine, decision tree, and random forest before and after using our approach. The findings showed that the proposed approach improved the predictive accuracy of the logistic regression, support vector machine, decision tree, and random forest algorithms in credit card datasets. Furthermore, we found that, out of the four algorithms, the random forest produced the best results.

Energy stability on sensor nodes in wireless sensor networks (WSNs) is always an important challenge, especially during data capturing and transmission of packets. The recent advancement in distributed clustering algorithms in the extant literature proposed for energy efficiency showed refinements in deployment of sensor nodes, network duration stability, and throughput of information data that are channelled to the base station. However, much scope still exists for energy improvements in a heterogeneous WSN environment. This research study uses the Gaussian elimination method merged with distributed energy efficient clustering (referred to as DEEC-Gauss) to ensure energy efficient optimization in the wireless environment. The rationale behind the use of the novel DEEC-Gauss clustering algorithm is that it fills the gap in the literature as researchers have not been able to use this scheme before to carry out energy-efficient optimization in WSNs with 100 nodes, between 1,000 and 5000 rounds and still achieve a fast time output. In this study, using simulation, the performance of highly developed clustering algorithms, namely, DEEC, EDEEC_E, and DDEEC, was compared to the proposed Gaussian Elimination Clustering Algorithm (DEEC-Gauss). The results show that the proposed DEEC-Gauss Algorithm gives an average percentage of 4.2% improvement for the first node dead (FND), a further 2.8% improvement for the tenth node dead (TND), and the overall time of delivery was increased and optimized when compared with other contemporary algorithms.

Social media networks such as Twitter are increasingly utilized to propagate hate speech while facilitating mass communication. Recent studies have highlighted a strong correlation between hate speech propagation and hate crimes such as xenophobic attacks. Due to the size of social media and the consequences of hate speech in society, it is essential to develop automated methods for hate speech detection in different social media platforms. Several studies have investigated the application of different machine learning algorithms for hate speech detection. However, the performance of these algorithms is generally hampered by inefficient sequence transduction. The Vanilla recurrent neural networks and recurrent neural networks with attention have been established as state-of-the-art methods for the assignments of sequence modeling and sequence transduction. Unfortunately, these methods suffer from intrinsic problems such as long-term dependency and lack of parallelization. In this study, we investigate a transformer-based method and tested it on a publicly available multiclass hate speech corpus containing 24783 labeled tweets. DistilBERT transformer method was compared against attention-based recurrent neural networks and other transformer baselines for hate speech detection in Twitter documents. The study results show that DistilBERT transformer outperformed the baseline algorithms while allowing parallelization.