Top-N movie recommendation method for performing weighted fusion on selected local models based on random anchor points

Abstract

The invention discloses a Top-N movie recommendation method for performing weighted fusion on selected local models based on random anchor points. The method comprises the steps of obtaining eigenvectors of users and movies at the semantic level through an LDA topic model and a GBDT by utilizing movie text data; based on the eigenvectors, calculating Gaussian kernel similarity between the users and the movies; randomly selecting a plurality of (the users and the movies) anchor point pairs, and reconstructing a local training matrix for each anchor point pair in combination with the Gaussian kernel similarity between the users and the movies; performing training for each local training matrix by utilizing an SLIM as a basic recommendation model to obtain a corresponding local recommendationmodel; and finally, generating a final fusion recommendation model through weighted fusion among the local recommendation models. According to the recommendation method, the stability of the models is also maintained in a data sparseness scene, and the problem that a traditional single recommendation model is very easy to over-fit in the data sparseness scene can be effectively solved.

Description

technical field [0001] The invention relates to a method for recommending movies on the Internet. Background technique [0002] With the development of Internet technology and socio-economic and cultural industries, more and more digital information such as electronic goods, digital news, online movies, and online videos appear on the Internet, and the implicit and explicit feedback data generated by users interacting with the Internet It also showed an exponential skyrocketing, and it became very difficult for users to find and find the information they are interested in from the massive data. The recommendation system can accurately predict user preferences based on the user's historical behavior information, help users quickly find the information they are interested in in massive data, and greatly improve the efficiency of information dissemination. [0003] Recommendation algorithms can be divided into content-based recommendation and collaborative filtering recommenda...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of overfitting in the single model recommendation algorithm of the prior art in the data sparse scene, the present invention draws on the idea of ​​integrated learning to obtain a strong classifier by training multiple weak classifiers, and provides a pair based on random anchor points. Selected Local Model Weighted Fusion Top-N Movie Recommendation Method

The recommendation method also maintains the stability of the model in the scenario of sparse data, and can effectively solve the problem that the traditional single recommendation model is prone to overfitting in the scenario of sparse data.

Images