Suspiciousness and Fast and Slow Thinking Impact on Trust in Recommender Systems

Abstract By offering consumers more proactive and individualized information services, recommender systems have proven to be a significant answer the problem of overload. And collaborative filtering approaches an important component many such systems, allowing for development high-quality recommendations by harnessing preferences communities similar users. In this study, we argue that individual personality traits may play essential role in ensuring general end-user confidence systems. Individual elements examined include social suspiciousness as feature Daniel Kahneman’s fast slow thinking styles, which distinguish between two modes thought: System 1 is rapid, instinctual, emotive, while 2 slower, deliberate, rational. Due lack explanation or inaccurate recommendation results, users not trust current poses challenge those As such, having trustworthy system essential. Three different trust-aware types are analyzed systematically study: robust filter misleading spam fake news; aware conscious benefit from users’ connections; explainable offer justifications recommended products. On sample size 487 Romanian respondents, online survey, impact factors on trustworthiness three Results show fully mediates relationship all