PERSONALIZED PRODUCT RECOMMENDATION USING PERSONALITY PROFILING AND META-PATH DISCOVERY

This paper introduces a novel method for personalized product recommendations in heterogeneous information networks by combining meta-path discovery with personality assessment. Neuroticism, agreeableness, extraversion, conscientiousness, and openness are among the user personality characteristics that the proposed system employs to establish a more comprehensive comprehension of individual preferences and behavioral patterns. Unlike traditional recommendation algorithms, this approach employs meta-path-based links in conjunction with psychological insights to identify more profound semantic connections between people, objects, and their environment. By incorporating both the user's expressed preferences and their concealed personality traits, the framework enhances the precision of its recommendations. The testing results indicate that the methodology significantly improves recommendation relevance, diversity, and user satisfaction in comparison to conventional collaborative filtering and content-based methods. In conclusion, the research demonstrates that the successful development of next-generation intelligent recommendation systems can be achieved by combining sophisticated network analysis with psychological modeling.