Multi-objective recommendation optimization method and readable medium

Abstract

The invention provides a multi-objective recommendation optimization method and a readable medium. The method comprises the following steps: initializing a plurality of decision vector families according to the number of to-be-recommended items and the number of recommended items; constructing a first objective function, a second objective function, a third objective function and a fourth objective function related to the decision vector families, wherein the first objective function is used for calculating the accuracy of the decision vectors, the second objective function is used for calculating the diversity of the decision vectors, the third objective function is used for calculating the novelty of the decision vectors, and the fourth objective function is used for calculating the coverage rate of the decision vector; constructing a final target function related to the decision vector family according to the first target function, the second target function, the third target function and the fourth target function; substituting the decision vector into the final target function to obtain a target function vector; comparing the target function vector to obtain a non-dominated solution set; generating a recommendation scheme of the to-be-recommended project according to the non-dominated solution set; and recommending the project to the target user according to the recommendation scheme.

Description

technical field [0001] The invention relates to the technical field of personalized recommendation, in particular to a multi-objective recommendation optimization method and a readable medium. Background technique [0002] With the development and maturity of Internet technology, the Internet is becoming more and more popular all over the world, which makes the amount of information on the network grow rapidly, and its growth rate reaches 2.5×10 per day. 18 byte. When users are faced with massive amounts of information, how to find the part that is really useful to them has become a very critical issue. Traditional information retrieval methods can no longer meet the needs of users, and personalized recommendation systems have emerged as the times require, and have been widely used in e-commerce, video sites, search engines and other fields, providing users with humanized, intelligent and convenient services. [0003] Among the existing personalized recommendation methods,...

AI Technical Summary

Problems solved by technology

The main goal of the collaborative filtering recommendation algorithm based on optimal clustering is to improve the data sparsity problem in the collaborative filtering algorithm, but in fact, the collaborative filtering algorithm becomes more complex when there are more users, and there are also problems such as "spoofing attacks" and poor scalability.

The content-based recommendation method uses machine learning to obtain user interest information from the characteristics of the project, and then examines the matching degree between the user and the project to be predicted and gives the recommendation result. The structural requirements for the feature content are high, and the recommendation result is excessive. The norm thus loses its novelty. At the same time, when a user with less historical data appears, it is difficult to give the ideal recommendation result

The recommendation method based on association rules is based on association rules. The purchased product is used as the rule head and the rule body as the recommendation object. The rule that satisfies the minimum support threshold and the minimum confidence threshold is regarded as a strong association rule for knowledge output. Algorithm The association rules of the first step are the most critical and time-consuming in the whole process

Since the scale of the relational database is generally very large, the complexity of the algorithm is extremely high, especially when the minimum support is low, a lot of frequent large itemsets need to be generated, the execution time of the algorithm will increase exponentially, and it is easy to cause hardware failure. memory exhausted

Images