Michael Cutter: Community Implicit and Explicit Feeback Recommender

Michael Cutter, a senior Information Systems Management major at the University of California, Santa Cruz, worked with Dr. Yi Zhang in the Information Retrieval Management laboratory of the University of California, Santa Cruz. Their research focused on search engine personalization and recommendations. Michael's work consisted of developing a platform that gathered implicit and explicit user feedback. The platform's primary function is to search user selected RSS (Really Simple Syndication) feeds and provide recommendations to users based on their history. The goal of this project is primarily to determine what browsing behavior implies relevance, and to provide a useful web application freely on the internet.

The platform consists of six components: A web application, a database server, a RSS parser, a RSS item indexer, a recommender system, and a Firefox add-on to capture implicit feed back.

In order to use this system you must make a login and choose some RSS feeds. The next step is to search the set of RSS feeds with the platform's search engine. Each term in a user's search query is checked against a list of words. If the word is a common piece of English such as "the" it is not logged. When the platform updates a RSS feed that is in the users set a query is performed with the logged keywords. If the article contains the keyword then it is flagged as a potential match with the user.

At the completion of this summer, Michael's platform was ready for the first set of user studies. The goal of these studies are to determine which implicit features determine interest in RSS feeds. Michael will continue to work on this project over the next year.