Yahoo! Movies User Ratings and Descriptive Content Information, v.1.0
Loading...
Files
Authors
Bodempudi, Ravitheja
Issue Date
2016-04-01
Type
Presentation
Language
en_US
Keywords
Data analysis , K-means clustering
Alternative Title
Abstract
A fundamental aspect of rating based systems is the observation process; the process which users choose the movies they rate. Finding user to user similarity is a fundamental component for collaborative filtering. In user to user similarity ratings assigned by two users to a set of items are pairwise compared and averaged is called correlation. In this project I want to show user to user similarity adaptive i.e., we dynamically change the computation depending on the profiles of the compared users and the target movie whose prediction is sought. I evaluate the proposed theory with k-means clustering by grouping similar users which rated similar movies with same rating. i.e., whoever is having same will come under one group.