An algorithm evaluation for discovering classification rules with gene expression programming

Abstract

In recent years, evolutionary algorithms have been used for classification tasks. However, only a limited number of comparisons exist between classification genetic rule-based systems and gene expression programming rule-based systems. In this paper, a new algorithm for classification using gene expression programming is proposed to accomplish this task, which was compared with several classical state-ofthe-art rule-based classifiers. The proposed classifier uses a Michigan approach; the evolutionary process with elitism is guided by a token competition that improves the exploration of fitness surface. Individuals that cover instances, covered previously by others individuals, are penalized. The fitness function is constructed by the multiplying three factors: sensibility, specificity and simplicity. The classifier was constructed as a decision list, sorted by the positive predictive value. The most numerous class was used as the default class. Until now, only numerical attributes are allowed and a mono objective algorithm that combines the three fitness factors is implemented. Experiments with twenty benchmark data sets have shown that our approach is significantly better in validation accuracy than some genetic rule-based state-of-the-art algorithms (i.e., SLAVE, HIDER, Tan, Falco, Bojarczuk and CORE) and not significantly worse than other better algorithms (i.e., GASSIST, LOGIT-BOOST and UCS).

GEP‐based classifiers with drift‐detection

“…Next, they designed and implemented a classifier system called GEPCLASS, producing flexible classification rules. An effective extension of the basic GEP classifier, named MCGEP, was recently proposed by Guerrero-Enamorado, Morell, Noaman, and Ventura (2016). The proposed classifier uses a Michigan approach, where each individual encodes a single classification rule.…”

Implementing Gene Expression Programming in the Parallel Environment for Big Datasets’ Classification

“…Usual mathematical operators were replaced by logical ones to calculate the chromosome quality. An e®ective extension of the basic GEP classi±er named MCGEP was recently proposed by Guerrero-Enamorado et al 14 The proposed classi¯er uses a Michigan approach where each individual encodes a single classi¯cation rule. Evolution process is based on elitism and the redundant individuals are penalized.…”

Multi-view Genetic Programming Learning to Obtain Interpretable Rule-Based Classifiers for Semi-supervised Contexts. Lessons Learnt

“…Therefore, multiple-induced rule-based systems are precisely subject of the multi-view approach, where the joint learning may produce better single systems even from the same dataset. A widely applied search paradigm to induce rule-based classifiers is genetic programming [22][23][24]. It applies the biological evolutionary metaphor on individuals that represent classification rules to generate robust classifiers.…”