A study of model to analyze the behavior of users of a personalized digital TV recommended system

摘要

In order to accurately describe and analyze human behavior. We need to build detailed, accurate user models. The models not only need to establish what users are interested in, but they also need to cover the details of the users' level of interest. Manual modeling or example modeling are not recommended because apparently these two methods will affect a user's normal programme selections. Research shows that users are normally reluctant to follow the recommended instructions, even though they are aware that by doing so the desired result will be achieved. What the users normally want to do is to find the desired programme with minimal effort and time. An automatic modeling method which is based on user's viewing behavior does not require the users active participation. Through the analysis of user's viewing behavior we can establish what users are interested in. This type of model is suitable for analyzing the use of programme recommended Systems for personalized TV.