A Movie Recommendation System—An Application of Voting Theory in User Modeling

Rajatish Mukherjee, Master’s Student, Computer Science, Partha Sarathi Dutta, Sandip Sen

Honorable Mention, Student Research Colloquium 2001

 

Our research agenda focuses on building software agents that can employ user-modeling techniques to solve day-to-day problems. Personal assistant agents embody a clearly beneficial application of intelligent agent technology. A particular kind of assistant agents, recommender systems, can be used to recommend items of interest to users. To be successful, such systems should be able to model and reason with user preferences for items in the application domain. Our primary concern is to develop a reasoning procedure that can meaningfully and systematically tradeoff between user preferences. We have adapted mechanisms from voting theory that have desirable guarantees regarding the recommendations generated from stored preferences. To demonstrate the applicability of our technique, we have developed a movie recommender system that caters to the interests of users. We present issues and initial results based on experimental data of our research that employs voting theory for user modeling, focusing on issues that are especially important in the context of user modeling. We provide multiple query modalities by which the user can pose unconstrained, constrained, or instance-based queries. Our interactive agent learns a user model by gaining feedback about its recommended movies from the user. We also provide proactive information gathering to make user interaction more rewarding. In my presentation, I will outline the current status of our implementation with particular emphasis on the mechanisms used to provide robust and effective recommendations.

 

Novelties of our work include:

 

1. Instance-based querying:

We have designed a combination of instance based learning and voting schemes by which the user can ask for a recommendation similar to a movie that he/she may have liked in the past. To provide this functionality, we have stored movies liked and disliked by the users in the past.

 

2. Learning:

We have incorporated text-based learning schemes by which the system can effectively update the stored user preferences based on recommendations that were liked or disliked by the user.

 

3. Proactive information gathering:

Whenever an existing user logs into the system, the system suggests a select set of newly released movies based on his/her previous selections. This enhances the user satisfaction level to a great degree.

 

Our movie recommendation system can be accessed at the following website: http://www.mcs.utulsa.edu/~rajatish/recommender/login.html