ONTOLOGY-BASED REAL-TIME SENTIMENT ANALYSIS FOR PRODUCT REPUTATION

Abstract

Social media platforms have become influential channels for user interaction, enabling the creation and exchange of content within virtual communities. E-commerce sites like Amazon, eBay, and CNET leverage user-generated reviews and ratings to influence purchasing decisions. However, current review systems are often centralized, static, and vulnerable to false evaluations, lacking the adaptability to reflect real-time consumer sentiment. This research proposes a dynamic, ontology-driven reputation system that continuously evolves to provide accurate and timely product ratings. The system collects online reviews, performs sentiment analysis to extract opinions, and uses a semantic layer to build contextual profiles for decision-making. An adaptive ontology integrates with a database to map relevant structures and generate semantic queries, delivering actionable insights. By enabling real-time adaptation and greater transparency, this approach empowers consumers and manufacturers to make informed decisions based on up-to-date reputational data