This study investigates the use of educational data mining (EDM) techniques to predict student performance and enhance learning outcomes in higher education. Leveraging data from Moodle, a widely used learning management system (LMS), we analyzed 450 students’ academic records spanning nine semesters. Five machine learning algorithms—k-nearest neighbors, random forest, logistic regression, decision trees, and neural networks—were applied to identify correlations between courses and predict grades. The results indicated that courses with strong correlations (+0.3 and above) significantly enhanced predictive accuracy, particularly in binary classification tasks. kNN and neural networks emerged as the most robust models, achieving F1 scores exceeding 0.8. These findings underscore the potential of EDM to optimize instructional strategies and support personalized learning pathways. This study offers insights into the effective application of data-driven approaches to improve educational outcomes and foster student success.

Maximum Relevance Minimum Redundancy (mRMR) is a widely used feature selection method that is applied in a wide range of applications in various fields. mRMR adds to the optimal subset the features that have high relevance to the target variable while having minimum redundancy with each other. Mutual information is a key component of mRMR as it measures the degree of dependence between two variables. However, the real value of mutual information is not known and needs to be estimated. The aim of this study is to examine whether the choice of mutual information estimator affects the performance of mRMR. To this end, three variations of mRMR are compared. The first one uses Parzen window estimation to assess mutual information between continuous variables. The second is based on equidistant partitioning using the cells method, while the third incorporates a bias-corrected version of the same estimator. All methods are tested with and without a regularization term in the mRMR denominator, introduced to improve numerical stability. The evaluation is conducted on synthetic datasets where the target variable is defined as a combination of continuous features, simulating both linear and nonlinear dependencies. To demonstrate the applicability of the proposed methods, we also include a case study in real-world classification tasks. The study carried out showed that the choice of mutual information estimator can affect the performance of mRMR and it must be carefully selected depending on the dataset and the parameters of the examined problem. The application of the corrected mutual information estimator improves the performance of mRMR in the examined setup.

This study presents an improved data-centric approach to short-term water demand forecasting using univariate time series from water reservoir levels. The dataset comprises water level recordings from 21 reservoirs in Eastern Thessaloniki collected over 15 months via a SCADA system provided by the water company EYATH S.A. The methodology involves data preprocessing, anomaly detection, data imputation, and the application of predictive models. Techniques such as the Interquartile Range method and moving standard deviation are employed to identify and handle anomalies. Missing values are imputed using LSTM networks optimized through the Optuna framework. This study emphasizes a data-centric approach in deep learning, focusing on improving data quality before model application, which has proven to enhance prediction accuracy. This strategy is crucial, especially in regions where reservoirs are the primary water source, and demand distribution cannot be solely determined by flow meter readings. LSTM, Random Forest Regressor, ARIMA, and SARIMA models are utilized to extract and analyze water level trends, enabling more accurate future water demand predictions. Results indicate that combining deep learning techniques with traditional statistical models significantly improves the accuracy and reliability of water demand predictions, providing a robust framework for optimizing water resource management.

This paper explores the application of parallel algorithms and high-performance computing (HPC) in the processing and forecasting of large-scale water demand data. Building upon prior work, which identified the need for more robust and scalable forecasting models, this study integrates parallel computing frameworks such as Apache Spark for distributed data processing, Message Passing Interface (MPI) for fine-grained parallel execution, and CUDA-enabled GPUs for deep learning acceleration. These advancements significantly improve model training and deployment speed, enabling near-real-time data processing. Apache Spark’s in-memory computing and distributed data handling optimize data preprocessing and model execution, while MPI provides enhanced control over custom parallel algorithms, ensuring high performance in complex simulations. By leveraging these techniques, urban water utilities can implement scalable, efficient, and reliable forecasting solutions critical for sustainable water resource management in increasingly complex environments. Additionally, expanding these models to larger datasets and diverse regional contexts will be essential for validating their robustness and applicability in different urban settings. Addressing these challenges will help bridge the gap between theoretical advancements and practical implementation, ensuring that HPC-driven forecasting models provide actionable insights for real-world water management decision-making.

Personalized learning environments increasingly rely on learner modeling techniques that integrate both explicit and implicit data sources. This study introduces a hybrid profiling methodology that combines psychometric data from an extended Felder–Silverman Learning Style Model (FSLSM) questionnaire with behavioral analytics derived from Moodle Learning Management System interaction logs. A structured mapping process was employed to associate over 200 unique log event types with FSLSM cognitive dimensions, enabling dynamic, behavior-driven learner profiles. Experiments were conducted across three datasets: a university dataset from the International Hellenic University, a public dataset from Kaggle, and a combined dataset totaling over 7 million log entries. Deep learning models including a Sequential Neural Network, BiLSTM, and a pretrained MLSTM-FCN were trained to predict student performance across regression and classification tasks. Results indicate moderate predictive validity: binary classification achieved practical, albeit imperfect accuracy, while three-class and regression tasks performed close to baseline levels. These findings highlight both the potential and the current constraints of log-based learner modeling. The contribution of this work lies in providing a reproducible integration framework and pipeline that can be applied across datasets, offering a realistic foundation for further exploration of scalable, data-driven personalization.

In feature selection, it is crucial to identify features that are not only relevant to the target variable but also non-redundant. Conditional Mutual Information Nearest-Neighbor (CMINN) is an algorithm developed to address this challenge by using Conditional Mutual Information (CMI) to assess the relevance of individual features to the target variable, while identifying redundancy among similar features. Although effective, the original CMINN algorithm can be computationally intensive, particularly with large and high-dimensional datasets. In this study, we extend the CMINN algorithm by parallelizing it for execution on Graphics Processing Units (GPUs), significantly enhancing its efficiency and scalability for high-dimensional datasets. The parallelized CMINN (PCMINN) leverages the massive parallelism of modern GPUs to handle the computational complexity inherent in sequential feature selection, particularly when dealing with large-scale data. To evaluate the performance of PCMINN across various scenarios, we conduct both an extensive simulation study using datasets with combined feature effects and a case study using financial data. Our results show that PCMINN not only maintains the effectiveness of the original CMINN in selecting the optimal feature subset, but also achieves faster execution times. The parallelized approach allows for the efficient processing of large datasets, making PCMINN a valuable tool for high-dimensional feature selection tasks. We also provide a package that includes two Python implementations to support integration into future research workflows: a sequential version of CMINN and a parallel GPU-based version of PCMINN.