Soft Genetic Programming Binary Classifiers

TL;DR

This paper introduces a 'soft' genetic programming approach for constructing binary classifiers, enhancing flexibility and dependency detection in datasets, with promising results demonstrated through tests.

Contribution

It presents a novel 'soft' genetic programming method for binary classification, addressing limitations of traditional GP classifiers and improving flexibility and dataset dependency detection.

Findings

Promising classification results demonstrated

Enhanced flexibility in logical operator trees

Effective dependency detection in datasets

Abstract

The study of the classifier's design and it's usage is one of the most important machine learning areas. With the development of automatic machine learning methods, various approaches are used to build a robust classifier model. Due to some difficult implementation and customization complexity, genetic programming (GP) methods are not often used to construct classifiers. GP classifiers have several limitations and disadvantages. However, the concept of "soft" genetic programming (SGP) has been developed, which allows the logical operator tree to be more flexible and find dependencies in datasets, which gives promising results in most cases. This article discusses a method for constructing binary classifiers using the SGP technique. The test results are presented. Source code - https://github.com/survexman/sgp_classifier.