Collaborative filtering algorithm based on user and project mixing

Abstract

The invention discloses a collaborative filtering algorithm based on user and project mixing, and the algorithm comprises the steps: 1, carrying out the arrangement of a user-project scoring data set, and building a user-project scoring matrix U; 2, calculating the similarities of articles, and ordering the similarities from the big to the small; 3, generating a 'nearest neighbor N' of articles according to the similarity ordering of the articles; 4, calculating the similarities between a target user T and other users, and ordering the similarities from the big to the small; 5, generating a 'nearest neighbor N' of users according to the similarity ordering of the users. The algorithm gives consideration to the similarities of the users and the similarities of the projects, obtains a project prediction score (giving consideration to the similarities of the users and the similarities of the projects at the same time) through employing a weighting method, carries out recommendation according to the ordering of scores, can reduce the value of an MAE (mean average error), and improves the accuracy of a recommended algorithm.

Description

technical field [0001] The invention relates to the technical field of personalized recommendation, in particular to a collaborative filtering algorithm based on mixing users and items. Background technique [0002] With the rapid development of network technology, people have begun to rely more and more on the Internet, through which people can query a large amount of information, but this also means that people have begun to enter the era of information overload. [0003] The information queried through the Internet is not necessarily what you really want to search for, so when faced with a large amount of data information, how to select the information that you are really interested in and useful to yourself has become a very difficult task matter. At the same time, for information producers, how to make their products stand out from a large number of product libraries, and find users who are really interested in them and willing to pay attention to them has become a ver...

AI Technical Summary

Problems solved by technology

With the continuous increase of projects and users, traditional algorithms will encounter serious scalability bottlenecks, which will affect the accuracy of collaborative filtering algorithms

Although the model-based algorithm can solve certain problems, the prerequisite for the model-based algorithm is that the user's hobbies remain basically unchanged, and this premise will also limit the use of the algorithm to a large extent.

[0010] Patent No. 201010613809.3 solves the above sparsity problem to a certain extent, but it solves the prediction accuracy problem by scoring matrix blanks and filling user-item matrix. The procedure is relatively complicated and the operation workload is heavy, especially when the data is sparse And in the case of a huge amount of data, the operability is poor

In addition, there will inevitably be certain errors in the process of filling the user-item matrix with this method, and the accuracy of the recommendation algorithm cannot be guaranteed

And it is not enough to solve the cold start and scalability problems of the existing technology

Images