Safeguarding Nations from Online News Threats Using Hybrid Technique

Abstract

Abstract: The internet provides a potent platform for individuals to express their opinions and emotions, facilitated by widespread smartphone usage and high internet accessibility. However, monitoring these online sentiments is crucial for identifying any extreme emotions that could potentially pose risks to national security. To address this, a new theoretical framework has been proposed, which combines a lexicon-based approach with machine learning techniques in the digital realm. This hybrid framework incorporates Decision Tree, Naive Bayes, and Support Vector Machine classifiers to predict political security threats. Through experimentation, it was found that the combination of a lexicon-based approach with the Decision Tree classifier yielded the highest performance score in predicting these threats. Natural Language Processing (NLP) techniques are employed for opinion mining within this framework