Research on Trust-Enhanced Recommender Algorithm Based on Weighting Approach

Compared to traditional algorithms that rely exclusively on user–item correlations, recommendation algorithms based on social networks, which incorporate users’ social data, can provide a more balanced approach to improving the precision of recommendations. This research introduces a trust-enhanced recommendation algorithm, called TECosRA, which integrates users’ historical behavior records with adjustable-weight social relationships by amplifying users’ social influence to enhance recommendation efficiency. Empirical results demonstrate that strengthening social connections and their positive impact on user decisions can broaden users’ choice channels and improve the accuracy of recommendation algorithms.