What is it about?
This paper reports on a study of the relationship between personality (using Big-five traits) and user retention and engagement, content preferences, and rating patterns for 1840 users in a recommender-system called MovieLens. We find that personality traits correlate significantly with behaviors and preferences such as newcomer retention, the intensity of engagement, activity types, item categories, consumption versus contribution, and rating patterns.
Featured Image
Photo by Sanket Kumar on Unsplash
Why is it important?
Recommender systems often lack the data they need to make the best possible recommendations to their users. Especially during the cold-start phase, algorithmic approaches for making recommendations face serious challenges because they lack sufficient (and diverse) information about user-item preferences. Moreover, recommender systems do not consider the varying tastes of users although they may capture user behavior. Furthermore, prior work suggests that mere improvements in accuracy do not necessarily provide better user experience, and that, more user-centric approaches should be considered for improving the experience of users in recommender systems. We, therefore, explore the value of personality in this work
Perspectives
More than three decades of research in social psychology and computer-mediated communication has shown personality to be correlated with user tastes, preferences, and behaviors in a wide variety of online and offline contexts. Personality plays a major role in decision-making and is found to be stable across the human lifespan. We, therefore, explore the potential of user personality in modeling user identity, activity, and preferences in the context of a recommender system.
Raghav Pavan Karumur
University of Minnesota System
Read the Original
This page is a summary of: Personality, User Preferences and Behavior in Recommender systems, Information Systems Frontiers, September 2017, Springer Science + Business Media,
DOI: 10.1007/s10796-017-9800-0.
You can read the full text:
Contributors
The following have contributed to this page
