Using Sentiment for User Profiling in Recommender Systems

Recommender systems rely on user profiling to generate personalized recommendations, typically using system usage data or explicitly expressed interests. This thesis explores the use of user sentiment as a profiling mechanism, integrating sentiment analysis into the recommendation process.

A prototype recommender system will be developed that utilizes sentiment to extract user profiles, with the aim of providing more personalized and contextually relevant recommendations. By incorporating emotional context into the profiling process, this approach seeks to enhance the accuracy and relevance of recommendations.