A method and system for cross-domain collaborative filtering

Abstract

The invention discloses a cross-domain collaborative filtering method. After converting the user item rating data into a training sample set, Funk-SVD decomposition is performed on the user item rating matrix of each auxiliary domain to obtain a user latent vector, and then the user latent vector is used. Vector extending the training sample set to obtain a first extended training sample set, adding item feature information to expand the first extended training sample set to obtain a second extended training sample set, using the second extended training sample set to train an unbalanced classifier, Finally, based on the unbalanced classifier, the missing data of the user item rating data is predicted and recommendations are generated; the problem of data sparsity in the target domain is solved by using auxiliary domain data to expand, and then the unbalanced classifier is trained on the expanded training samples , using the unbalanced classifier to predict the missing items in the target domain, and then get the recommended data, which solves the sparse and unbalanced problems of the existing recommendation system data sets.

Description

technical field [0001] The invention belongs to the technical field of information recommendation, and in particular relates to a cross-domain collaborative filtering method and system. Background technique [0002] The rapid growth of Internet information requires effective intelligent information agents that can filter out all available information and find the most valuable information for users. [0003] In recent years, recommendation systems have been widely used in e-commerce networks and online social media. At present, the main recommendation methods are divided into: content-based recommendation, collaborative filtering-based recommendation, association rule-based recommendation, utility-based recommendation, knowledge-based recommendation, Combination recommendation, etc. Among them, the recommendation based on collaborative filtering is the most successful strategy in the recommendation method. The basic idea is that the user is likely to like the resource that i...

AI Technical Summary

Problems solved by technology

[0005] This application provides a cross-domain collaborative filtering method and system to solve the technical problem of unbalanced data sets in existing recommendation systems

Images