Classification is an important topic in data mining research. Given a set of data records, each of which belongs to one of a number of predefined classes, the classification problem is concerned with the discovery of classification rules that can allow records with unknown class membership to be correctly classified. Many algorithms have been developed to mine large data sets for classification models and they have been shown to be very effective. However, when it comes to determining the likelihood of each classification made, many of them are not designed with such purpose in mind. For this, they are not readily applicable to such problem as churn prediction. For such an application, the goal is not only to predict whether or not a subscriber would switch from one carrier to another, it is also important that the likelihood of the subscriber's doing so be predicted. The reason for this is that a carrier can then choose to provide special personalized offer and services to those subscribers who are predicted with higher likelihood to churn. Given its importance, we propose a new data mining algorithm, called data mining by evolutionary learning (DMEL), to handle classification problems of which the accuracy of each predictions made has to be estimated. In performing its tasks, DMEL searches through the possible rule space using an evolutionary approach that has the following characteristics: 1) the evolutionary process begins with the generation of an initial set of first-order rules (i.e., rules with one conjunct/condition) using a probabilistic induction technique and based on these rules, rules of higher order (two or more conjuncts) are obtained iteratively; 2) when identifying interesting rules, an objective interestingness measure is used; 3) the fitness of a ch...

The paper proposes a linear representation of tree structures in order to evolve complex fuzzy rule sets for solving classification problems. In particular, linguistic rules are evolved, where the condition part of a rule can have an arbitrary structure of conjunctions and disjunctions. We describe an efficient rule representation scheme, which uses a genetic algorithm. The method is tested with a number of benchmark data sets and some results are reported.

This work is intended to discover classification rules for diagnosing certain pathologies. In order to discover these rules we have developed a new constrained-syntax genetic programming algorithm based on some concepts of data mining, particularly with emphasis on the discovery of comprehensible knowledge. We compare the performance of the proposed GP algorithm with a genetic algorithm and with the very well-known decision-tree algorithm C4.5.

We propose a new linear representation scheme for evolving fuzzy rules using the concept of complete binary tree structures. We also use special genetic operators such as gene addition, gene deletion, and variable length crossover. Results show that using these special operators along with the common mutation operator produce useful and minimal structure modifications to the fuzzy expression tree represented by the chromosome. The proposed method (representation and operators) is tested with a number of benchmark data sets including the KDDCup'99 Network Intrusion Detection data.

Abstract—In the recent past, there has been an increasing interest in applying evolutionary methods to Knowledge Discovery in Databases (KDD) and a number of successful applications of Genetic Algorithms (GA) and Genetic Programming (GP) to KDD have been demonstrated. The most predominant representation of the discovered knowledge is the standard Production Rules (PRs) in the form If P Then D. The PRs, however, are unable to handle exceptions and do not exhibit variable precision. The Censored Production Rules (CPRs), an extension of PRs, were proposed by Michalski & Winston that exhibit variable precision and supports an efficient mechanism for handling exceptions. A CPR is an augmented production rule of the form: If P Then D Unless C, where C (Censor) is an exception to the rule. Such rules are employed in situations, in which the conditional statement ‘If P Then D ’ holds frequently and the assertion C holds rarely. By using a rule of this type we are free to ignore the exception conditions, when the resources needed to establish its presence are tight or there is simply no information available as to whether it holds or not. Thus, the ‘If P Then D ’ part of the CPR expresses important information, while the Unless C part acts only as a switch and changes the polarity of D to ~D. This paper presents a classification algorithm based on evolutionary approach that discovers comprehensible rules with exceptions in the form of CPRs. The proposed approach has flexible chromosome encoding, where each chromosome corresponds to a CPR. Appropriate genetic operators are suggested and a fitness function is proposed that incorporates the basic constraints on CPRs. Experimental results are presented to demonstrate the performance of the proposed algorithm.

Automated discovery of hierarchical structures in large data sets has been an active research area in the recent past. This paper focuses on the issue of mining generalized rules with crisp hierarchical structure using Genetic Programming (GP) approach to knowledge discovery. The post-processing scheme presented in this work uses flat rules as initial individuals of GP and discovers hierarchical structure. Suitable genetic operators are proposed for the suggested encoding. Based on the Subsumption Matrix(SM), an appropriate fitness function is suggested. Finally, Hierarchical Production Rules (HPRs) are generated from the discovered hierarchy. Experimental results are presented to demonstrate the performance of the proposed algorithm.

A number of soft computing approaches, such as neural networks, evolutionary algorithms, and fuzzy logic, have been widely used for classifier agents to adaptively evolve solutions on classification problems. However, most work in the literature focuses on the learning ability of individual classifier agent. This paper explores incremental, collaborative learning in a multi-agent environment. We use genetic algorithm (GA) and incremental genetic algorithm (IGA) as the main techniques to evolve the rule set for classification, and employ new class acquisition as a typical example to illustrate the incremental, collaborative learning capability of classifier agents. Benchmark data sets are used to evaluate proposed approaches. The results show that GA and IGA can be successfully used for collaborative learning among classifier agents.

Genetic algorithms (GAs) have been widely used as soft computing techniques in various applications, while cooperative co-evolution algorithms were proposed in the literature to improve the performance of basic GAs. In this paper, a new cooperative co-evolution algorithm, namely ECCGA, is proposed in the application domain of pattern classification. Concurrent local and global evolution and conclusive global evolution are proposed to improve further the classification performance. Different approaches of ECCGA are evaluated on benchmark classification data sets, and the results show that ECCGA can achieve better performance than the cooperative co-evolution genetic algorithm and normal GA. Some analysis and discussions on ECCGA and possible improvement are also presented.

Abstract. In this paper we address the problem of multiobjective attribute selection in data mining. We propose a multiobjective genetic algorithm (GA) based on the wrapper approach to discover the best subset of attributes for a given classification algorithm, namely C4.5, a well-known decision-tree algorithm. The two objectives to be minimized are the error rate and the size of the tree produced by C4.5. The proposed GA is a multiobjective method in the sense that it discovers a set of non-dominated solutions (attribute subsets), according to the concept of Pareto dominance. 1

Abstract — This paper presents a framework for genetic fuzzy rule based classifier. First, a classification problem is divided into several two-class problems following a fuzzy class binarization scheme; next, a fuzzy rule is evolved for each two-class problem using a Michigan iterative learning approach; finally, the evolved fuzzy rules are integrated using the fuzzy class binarization scheme. In particular, some encoding schemes are implemented following the proposed framework and their performance is compared. Experiments are conducted with different public available data sets.