A television program hybrid recommendation method based on an MCL-HCF algorithm

Abstract

The invention belongs to the technical field of television program recommendation, and discloses a television program hybrid recommendation method based on an MCL-HCF algorithm, which comprises the following steps of: firstly, clustering television users in each time period by adopting Markov clustering to generate different groups, pursuing to minimize the preference difference between members in each group and the whole body of a group owner, and then recommending television programs by taking the group as a unit; Then, respectively generating recommendation lists by using an article-basedcollaborative filtering algorithm and a user-based collaborative filtering algorithm; And finally, in order to balance the surprising degree and the correlation of the recommendation results, mixing the two recommendation lists in a weighting manner to obtain a final mixed recommendation result. According to the invention, the preference difference between users in the group and the whole group isreduced; the results of the two recommended algorithms including ItemCF-IUF and UserCF-IIF are weighed and mixed to solve the problem of contradiction between the surprise degree and the correlationof the recommendation results; And while the recommendation accuracy is kept, the surprising degree and the correlation of the recommended programs are balanced.

Description

technical field [0001] The invention belongs to the technical field of TV program recommendation, and in particular relates to a TV program hybrid recommendation method based on an MCL-HCF algorithm. Background technique [0002] Currently, the existing technology commonly used in the industry is this: Watching TV programs has always been an important part of human spiritual life since the birth of television. Today, due to the rapid development of computer technology and network technology, people are more and more accustomed to watching video programs on the Internet platform, which has also brought an impact on traditional broadcasting and TV operators. For broadcasters, the loss of customers has brought many challenges, but also new opportunities. Premium channels are now the bread and butter of broadcast television and a significant source of revenue. The problem in the existing technology is: in the recommendation stage, the traditional item-based collaborative filte...

AI Technical Summary

Problems solved by technology

[0003] To sum up, the problem existing in the existing technology is: in the recommendation stage, the traditional item-based collaborative filtering and user-based collaborative filtering do not take into account the impact of user activity and item audience degree on similarity calculation, affecting the final Recommendation effect; ItemCF-IUF recommends similar items to users, so the surprise degree of the programs it recommends is relatively low; UserCF-IIF recommends based on user similarity, and the correlation of recommendation results is relatively weak

Images