Contract machine recommendation method and recommendation system

Abstract

The invention discloses a contract machine recommendation method and recommendation system in the technical field of machine learning. In the technical scheme provided by the invention, the feature vector of the current user and the feature vector of each contract machine are input into the factorization machine model and the logistic regression model, the probability that the current user orders each contract machine is obtained, and a contract machine recommendation list is generated, so that the contract machines are recommended to the user, and the accuracy of recommending the contract machine to the user is improved. Furthermore, the feature information of the contract machines which are purchased by the target user in history is used as the feature information of the contract machines which are purchased by the user in history and do not have the contract machine history purchase information, and the target user is the user who has the largest similarity with each user who does not have the contract machine history purchase information and has the contract machine history purchase information. The factorization machine model and the logistic regression model are trained, so that model training samples are enriched, and the accuracy of the factorization machine model and the logistic regression model is improved.

Description

technical field [0001] This application relates to the technical field of machine learning, and in particular to a recommendation method and system for a contract machine. Background technique [0002] Contract phone generally refers to the legal mobile phone type customized by communication operators (such as China Mobile, China Unicom, China Telecom, etc.) and mobile phone manufacturers. To use the contract machine, the user must sign a contract with a designated communication operator and use a designated communication package. [0003] Compared with bare phones (phone buyers can choose their own communication operators and can freely choose packages), contract phones are more in line with the needs of users in terms of equipment prices and package content. [0004] With the development of communication technology, the functions of contract machines are becoming more and more complete and rich, and there are more and more types. Therefore, how to accurately recommend su...

AI Technical Summary

Problems solved by technology

[0006] However, the above recommendation method has the following problems: there are many types of contract machines, and it is impossible to accurately recommend suitable contract machines to users.

Images