tagommendersconnectinguserstoitemsthroughtags连接用户通过标签tagommenders的项目.ppt

Tagommenders : Connecting Users to Items through Tags Written by Shilad Sen, Jesse Vig, and John Riedl (2009) Presented by Ken Hu and Hassan Hattab Overview a.Introduction b.Tagommender a. Philosophy b. Dataset c.Tag Preference Inference a. Approach b. Methods a.Recommenders a. Implicit b. Explicit b.Results c.Conclusion First, Recommenders. a.What is Recommender system? b.Two Main tasks a. Recommend. b. Predict. Recommender Systems a.Types of recommender systems: a. User-based : decides according to the user's previous choices b. Item-based : decides according to related items to a selected item c. SVD ? Problem: These methods don't consider the content of the item. ? Solution : Content-based Recommenders Overview a.Introduction b.Tagommender a. Philosophy b. Dataset c.Tag Preference Inference a. Approach b. Methods a.Recommenders a. Implicit b. Explicit b.Results c.Conclusion Tagging Systems a.Uses tags to address (categorize) items to users b.Tags are created by general users ( More meaningful ) Tagommenders: a.Basically, they combine Recommenders (content- based) and tagging systems. ? Two main parts for Tagommenders: o They infer users' preferences for tags based on their interactions with tags and movies o and they infer users' preferences for movies based on their preferences for tags. Tagommender's data set These are collected from the MovieLens website. a.Movie Rating ? Movie clicks ? Tag applications ? Tag Searches ? Tag Preference Ratings Tagommender's data set Overview a.Introduction b.Tagommender a. Philosophy b. Dataset c.Tag Preference Inference a. Approach b. Methods a.Recommenders a. Implicit b. Explicit b.Results c.Conclusion Tagommender's Cycle Inferring Tag Preference a. Inferring Preference using Tag Signals (Direct) Inferring Tag Preference a. Inferring Preference usin