1 / 7

Active Collaborative Filtering

Active Collaborative Filtering. Machine Learning Group Department of Computer Science University of Toronto. Users express preference for items they have viewed, accessed, or purchased by assigning ratings to them.

azriel
Télécharger la présentation

Active Collaborative Filtering

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Active Collaborative Filtering Machine Learning Group Department of Computer ScienceUniversity of Toronto

  2. Users express preference for items they have viewed, accessed, or purchased by assigning ratings to them. Collaborative filtering systems analyze the preference data to make customizedrecommendationsandpredictions for each user. Collaborative Filtering:

  3. When a new user first joins a collaborative filtering system their rating profile is empty, and recommendations can be of poor quality. This is often called the New User Problem, and itaffects all collaborative filtering systems. Our approach to Active Collaborative Filtering applies principled methods from decision theoryto help overcome the new user problem by guiding the rating process. The Active Advantage:

  4. Recent research has shown that our approach to ACF provides a significant improvement over entering ratings in a haphazard fashion. It also outperforms other methods that have been proposed in the past. Proven Results:

  5. Proven Results: Improvement in Recommendation Quality (MCVQ)

  6. Proven Results: Improvement in Recommendation Quality (NB)

  7. Active Movie Recommendation Demo Includes 115 titles. Fully interactive in real time. Use the active query option or enter ratings manually. Top five list automatically recalculated.

More Related