A new project collaborative recommendation method based on multi-core fusion

Abstract

The invention relates to a cold start recommendation algorithm based on commodity attribute information, and aims to solve the problem of data loss in new commodity recommendation by using a multi-core weighted fusion collaborative filtering algorithm. According to the algorithm, the incidence relation of the commodities in the attribute space is determined in a multi-core weighting mode, and therefore new projects are recommended to the user. Wherein the multi-kernel learning algorithm is based on an existing kernel function learning algorithm and is used for carrying out weighted summation on all kernel functions, so that the accuracy of the algorithm in a complex data environment is improved; Wherein the attribute similarity is obtained by calculating the similarity of attributes amongthe commodities, so that the calculated preference score of the user for commodity prediction is more interpretable; Wherein the weight is optimized through a learning method of random gradient descent. According to the method and the device, the item similarity measurement for describing user preferences can be learned according to the attribute information of the commodities, so that the new item recommendation accuracy is effectively improved.

Description

technical field [0001] The invention relates to the field of machine learning, in particular to a new item collaborative recommendation method based on multi-core fusion. Background technique [0002] With the rapid development of social economy, people will be faced with more and more commodities, information and services, causing users to have no choice. In order to solve this problem of information overload, recommender systems came into being. The most widely used is the item-based collaborative filtering algorithm. Its core idea is to recommend similar items to the user that he (she) liked before, and can perform personalized recommendations without requiring other user information. However, if the historical evaluation information of items is missing, this method will be difficult to generate effective recommendations, which is the new item recommendation problem (also known as the item cold start problem). [0003] There are three commonly used cold start recommenda...

AI Technical Summary

Problems solved by technology

[0004] (1) Use the scoring median, mean, mode, etc. to fill in the gaps, but the scoring values ​​of unrated items will be exactly the same, so the credibility of this method is not high;

[0005] (2) Using the method of nearest neighbor to calculate the similarity between items, but most of the nearest neighbor items found have not been rated by the current user, so the stability is not good;

[0006] (3) Using the similarity method to calculate the similarity between items, but because it is difficult to achieve an effective match between the similarity and user preferences, the adaptability of the algorithm is not strong

Images