High-quality mode mining method based on multi-objective evolutionary algorithm

Abstract

The invention discloses a high-quality mode mining method based on a multi-objective evolutionary algorithm, and belongs to the technical field of data mining. According to the method, the problem that it is difficult for a user to set an appropriate parameter threshold value is solved by adopting a multi-objective evolutionary algorithm; OR / NOR-tree structure based population initialization strategy is combined with an original database which has been represented as a bitmap form to construct an initial population, and an improved cross operator and a mutation operator are adopted to set theNOR position and OR position in the OR / NOR-tree structure, so that the problem that the traditional random initialization method and crossover and mutation operator efficiency are not high due to thefact that the data is usually huge and sparse is solved; in addition, a worst individual search direction adjustment strategy is adopted to adjust the search direction, the optimization process and the optimization result are improved, and the convergence speed and the final solution quality are improved.

Description

technical field [0001] The invention relates to a high-quality pattern mining method based on a multi-objective evolutionary algorithm, and belongs to the technical field of data mining. Background technique [0002] Data mining refers to the process of extracting potentially interesting information or patterns from a large amount of data for further use. [0003] Most of the traditional pattern mining methods need to set a priori parameters. For users without any experience, how to set an appropriate parameter threshold is difficult, and the solution efficiency is relatively low. However, the multi-objective evolutionary algorithm can explore the patterns that meet the specified conditions without setting the threshold. [0004] Existing multi-objective pattern mining algorithms, such as Pattern Recommendation in Task-oriented Applications: A Multi-Objective Perspective published in 2017, transform the task-oriented pattern mining problem into a multi-objective optimizatio...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the solution efficiency of the existing data mining model algorithm is low, the present invention provides a high-quality pattern mining method based on the multi-objective evolutionary algorithm, the method is based on the NSGA-II algorithm, and the following steps are adopted in the algorithm It improves:

Images