Comparative Study of Ensemble and Neural Models for Insider Threat Detection Under Class Imbalance (Published)
Insider threats remain one of the most difficult cybersecurity risks to detect because malicious activities often originate from legitimate users operating within authorised boundaries. Machine learning techniques have increasingly been applied to insider threat detection; however, there is limited empirical evidence comparing the effectiveness of classical machine learning models and deep learning architectures on large-scale behavioural datasets under realistic class imbalance conditions. This study presents a comparative performance evaluation of ensemble machine learning and neural deep learning models for insider threat detection using a large publicly available behavioural risk dataset comprising 299,880 employee activity records. After rigorous preprocessing, feature engineering, and class balancing through controlled undersampling, four models were evaluated: Random Forest, Extreme Gradient Boosting (XGBoost), Multi-Layer Perceptron (MLP), and an Autoencoder-enhanced MLP (AE-MLP). Experimental results show that ensemble tree-based methods outperform deep neural models on tabular behavioural data, with XGBoost achieving the best overall performance (Accuracy 0.894, F1-score 0.895, ROC-AUC 0.969). Deep learning models demonstrated competitive precision but lower recall, indicating reduced sensitivity to malicious behaviour patterns. To validate model behaviour, SHAP-based global feature importance analysis was applied to the best-performing model, confirming that predictions relied on meaningful behavioural indicators, including data transfer activity, printing behaviour, access timing, and employee role characteristics. The findings suggest that for structured insider threat datasets, optimised classical ensemble models remain more effective and computationally efficient than deep neural approaches, while lightweight explainability methods can provide useful behavioural validation without heavy interpretability overhead.
Keywords: Comparative study, Ensemble, insider threat detection under class imbalance, neural models
Melody Analysis for Prediction of the Emotions Conveyed by Sinhala Songs (Published)
This paper describes our attempt of assessing the capability of music melodies in isolation in order to classify music files into different emotional categories in the context of Sri Lankan music. In our approach, Melodies (predominant pitch sequences) are extracted from songs and the feature vectors are created from them which are ultimately subjected to supervised learning approaches with different classifier algorithms and also with classifier accuracy enhancing algorithms. The models we trained didn’t perform well enough to classify songs into different emotions, but they always showed that the melody is an important factor for the classification. Further experiments with melody features along with some non-melody features showed us that those feature combinations perform much better, hence brought us to the conclusion that, even though, the melody plays a major role in differentiating the emotions into different categories, it needs the support of other features too for a proper classification.
Keywords: Emotion Classification, Ensemble, Feature Selection, Melody, Music Information Retrival, Supervised Learning