Cyberbullying via social media is a constant digital safety issue because the content can be widely shared and openly visible and can have a negative impact on users before it is removed by manual moderation. Current detection models are mostly based on shallow lexical features or transformer-only classifiers, resulting in low-level accuracy and explainability. This study introduces a Hybrid BERT–XGBoost model to detect cyberbullying in short social media texts, which combines the strengths of both models. The contextual sentence embeddings are extracted using BERT and the auxiliary linguistic and behavioral features are extracted in parallel, such as sentiment polarity, profanity score, punctuation intensity, capitalization ratio, hashtag usage, mention count, emoji frequency, and post length. XGBoost is used for the classification of the fused representation. The model is tested on stratified training, validation, and testing splits, compared to a baseline model, ablated, tested with macro-F1, weighted-F1, ROC-AUC, early detection recall, and grouped explainability. The proposed framework achieved 96.18% accuracy, 96.05% macro-F1, 96.16% weighted-F1, 95.88% early detection recall, and 98.42% macro-AUC. It performs better than the BERT + Dense baseline, which obtained 94.31% accuracy and 94.08% macro-F1 score, demonstrating the advantage of fusion of contextual and auxiliary features. The framework provides an interpretable, practical and category-aware solution for early detection of cyberbullying, but further research is needed to validate the framework in multiple languages, modalities and in conversations.
Rima Shishakly, Abdelhadi Mohammad Bayoud, Mansour Obeidat et al.· Journal of Cyber Security an...· 0 citations
Reliable short-term PV forecasting is essential to ensure the reliability of energy management based on renewable sources, but the lack of interpretability of deep learning models and the environmental uncertainty still prevent their operation. In this research, an uncertainty-aware transformer framework is proposed which is explainable when forecasting the short-term PV power under dynamically changing environmental conditions. It introduces a framework which combines a Temporal Fusion Transformer (TFT), quantile-based uncertainty estimation, dual explainability mechanism based on attention analysis, and SHAP based feature attribution. The measurements from a real solar system having capacity of 1.005 kW were taken at the site of Sitapura, Jaipur, India and were synchronously matched with the POWER meteorological measurements taken by NASA. The framework was tested on the baseline models of ARIMA, LSTM and CNN-LSTM. These experimental results showed that the approach achieved better forecasting results compared to all the baseline approaches with RMSE of 0.067, MAE of 0.051 and MAPE of 4.38%. The probabilistic forecasting module was able to attain a Prediction Interval Coverage Probability (PICP) of 94.8% and narrow interval width, thus demonstrating good uncertainty estimation abilities. Irradiance and PV output of the past were found to be the most important factors for forecasting in this explainability analysis. The proposed framework is based on the transformer model, probabilistic forecasting, and then explains the renewable energy forecasting system in the entire system for photovoltaic uncertainty-aware and interpretable prediction.
Udit Mamodiya, I. Kishor, P. Mudholkar et al.· International Journal of Dat...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
