Did you know that your personality type can be used to predict your behavior on an online recommender site (how long you stay, what you do, whether and how much you are likely to rate)  and even what to recommend to you? That’s what we found in our latest research using the MovieLens recommender system and the Big Five Personality scale for modeling user personality. To learn more, read on!

In our first study, we examined newcomer retention, type, and level of contribution associated with various personality types for 1008 new users. We found that if you are an introvert, you are likely to stay longer and be more active on MovieLens. If you are highly open, you are likely to tag more, and if you are less agreeable, you are likely to provide more ratings!!

In our next study, we looked at how users with different personality types vary in their preferences for various categories such as movie Genres, Popularity (high or low), and Rating level (low-rated movies, medium-rated, and high-rated).

As the picture above shows, highly agreeable users rate movies across a number of categories about half-star higher than less agreeable ones.

Findings such as these suggest that aggregating ratings of all users to provide recommendations may not be a good idea. Also, documenting stable relationships between individual differences and category preferences can be useful for online system designers as they customize the system to guide people towards tasks that need to be done (such as recommending the right categories to elicit ratings from users), or tasks the users will find rewarding (such as providing better cold-start recommendations, personalizing lists of movies at different times) and also decide which users to invest retention efforts in. 

To know more about our data collection process, analyses, and for a full list of all of our findings on various personality types, please check out our papers on modeling newcomer personality, personality and category preferences (video), and personality and user behavior on recommender systems!

Written by

Raghav Pavan Karumur is a Ph.D. Candidate at GroupLens Research. Broadly, his research interests include recommender systems, personalization, user modeling, social computing, applied machine learning, and feature engineering. He applies concepts from social science, behavioral science, and organizational theory to understand individual as well as crowd behaviors in these communities to personalize and design better systems for users, recommend items matching their preferences, and devise strategies that increase their commitment and motivate contributions in these systems. He was a member of IEEE and is currently a member of ACM. He has published in top-tier conferences in Human-Computer Interaction such as CHI, CSCW, UMAP and RecSys and top-tier journals such as IJHCI and Information Systems Frontiers. He has reviewed and has served on the Program Committee for conferences and journals such as UMAP, RecSys and CHI, DIS, and Journal of Intelligent Information Systems.

Comments are closed.