Using Genetic Algorithms for Concept Learning

Jong, Kenneth A. De; Spears, William M.; Gordon, Diana F.

doi:10.1007/978-1-4615-2740-4_2

Cited by 122 publications

(173 citation statements)

References 1 publication

Supporting

Mentioning

164

Contrasting

Unclassified

Order By: Relevance

“…As we commented in Sect. 4.2, in the specialized literature other proposals have been considered, such as codifying within the individual's genome [17] (in this way consequent evolution is also possible) or the deterministic selection for each rule of the value of the target feature [27,49]. In problems such as the one we present, the focus we have adopted is suitable because it is necessary to describe all the values of the target feature, and the two alternatives mentioned above do not ensure information extraction relating to all the classes.…”

Section: Chromosome Representationmentioning

confidence: 99%

Evolutionary Induction of Descriptive Rules in a Market Problem

Jesus

González

Herrera

et al. 2005

Intelligent Data Mining

View full text Add to dashboard Cite

Abstract. Nowadays, face to face contact with the client continues to be fundamental to the development of marketing acts. Trade fairs are, in this sense, a basic instrument in company marketing policies, especially in Industrial Marketing. Due to the elevated investment in term of both time and money it is necessary the automatic extraction of relevant and interesting information which helps to improve fair planning policies.In this paper, we analyse this problem and the kind of knowledge the user is interested in. We study the use of Soft Computing methodologies, specifically Fuzzy Logic and Genetic Algorithms, in the design of the Data Mining algorithms most proper to this problem, descriptive induction algorithms for subgroup discovery. Then we present an evolutionary model for the descriptive induction of fuzzy or crisp rules which describe subgroups. The proposal includes a GA in an iterative model which extracts rules while examples are left uncovered and the rules obtained surpass an specified confidence level.

show abstract

Section: Chromosome Representationmentioning

confidence: 99%

Evolutionary Induction of Descriptive Rules in a Market Problem

Jesus

González

Herrera

et al. 2005

Intelligent Data Mining

View full text Add to dashboard Cite

show abstract

“…Genetic Algorithms (GA) have been shown to be an effective tool to use in data analysis and pattern recognition [1], [2], [3]. An important aspect of GAs in a learning context is their use in pattern recognition.…”

Section: Background On Using Gas For Feature Selection/extractionmentioning

confidence: 99%

Optimizing Classification Ensembles via a Genetic Algorithm for a Web-Based Educational System

Minaei‐Bidgoli

Kortemeyer

Punch

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Classification fusion combines multiple classifications of data into a single classification solution of greater accuracy. Feature extraction aims to reduce the computational cost of feature measurement, increase classifier efficiency, and allow greater classification accuracy based on the process of deriving new features from the original features. This paper represents an approach for classifying students in order to predict their final grades based on features extracted from logged data in an educational web-based system. A combination of multiple classifiers leads to a significant improvement in classification performance. By weighing feature vectors representing feature importance using a Genetic Algorithm (GA) we can optimize the prediction accuracy and obtain a marked improvement over raw classification. We further show that when the number of features is few, feature weighting and transformation into a new space works efficiently compared to the feature subset selection. This approach is easily adaptable to different types of courses, different population sizes, and allows for different features to be analyzed.

show abstract

“…Classification functions studied range from a set of data-attribute weights as in traditional regression models [5] [6], condition-action type rules with conjunction/disjunction of terms [7] [8], to the parse-tree expressions of genetic programming [9]. The flexibility in fitness function formulation allows the development of classification models that are tailored to specific domain constraints and objectives.…”

Section: Introductionmentioning

confidence: 99%