Method for solving minimum attribute reduction by combining local opponent learning and social spider algorithm

Abstract

The invention discloses a method for solving minimum attribute reduction by combining local opponent learning and a social spider algorithm, which comprises the following steps: proposing a similarityconstraint at the beginning of iteration to keep individuals in a population in a good state, introducing opponent learning in the iteration process, designing a local opponent learning strategy to expand the search range, and optimizing the search result; increasing the convergence speed; and furthermore, adopting a redundancy detection mechanism to carry out redundancy detection on the globallyoptimal solution to guarantee minimum attribute reduction as much as possible. According to the method, effective minimum reduction can be found under most conditions, the operation time is shorter,and meanwhile the convergence speed is higher; and along with the increase of a data set, a better performance can also be shown.

Description

technical field [0001] The invention relates to the field of data mining and knowledge discovery, in particular to a method for solving minimum attribute reduction by combining local adversarial learning and social spider algorithm. Background technique [0002] With the development of society, the scale of data increases exponentially, and a large number of noisy, irrelevant or misleading features are generated in the data. Rough Set Theory (RST) is a mathematical tool for dealing with uncertain, imprecise and fuzzy data. Rough sets are widely used in many fields, such as machine learning, feature selection, data mining, image processing, pattern recognition. RST can remove redundant attributes in data by discovering dependencies in data without any prior information. Given a dataset with discretized values, a subset (reduction) of the original set can be found via RST. At present, many scholars have conducted extensive research in the field of attribute reduction. Amon...

AI Technical Summary

Problems solved by technology

But in practice, it is unrealistic to generate all reductions, and the minimum attribute reduction problem is an NP-hard problem

However, the introduction of an adversarial learning mechanism for intelligent algorithms is usually accompanied by a large amount of computational redundancy, and with the increase of individuals in the intelligent algorithm and the increase in the number of iterations, the amount of calculation generated by adversarial learning seriously affects the running time.

Images