A hybrid decision tree/genetic algorithm method for data mining

Abstract

This paper addresses the well-known classification task of data mining, where the objective is to predict the class which an example belongs to. Discovered knowledge is expressed in the form of high-level, easy-to-interpret classification rules. In order to discover classification rules, we propose a hybrid decision tree/genetic algorithm method. The central idea of this hybrid method involves the concept of small disjuncts in data mining, as follows. In essence, a set of classification rules can be regarded as a logical disjunction of rules, so that each rule can be regarded as a disjunct. A small disjunct is a rule covering a small number of examples. Due to their nature, small disjuncts are error prone. However, although each small disjunct covers just a few examples, the set of all small disjuncts can cover a large number of examples, so that it is important to develop new approaches to cope with the problem of small disjuncts. In our hybrid approach, we have developed two genetic algorithms (GA) specifically designed for discovering rules covering examples belonging to small disjuncts, whereas a conventional decision tree algorithm is used to produce rules covering examples belonging to large disjuncts. We present results evaluating the performance of the hybrid method in 22 real-world data sets. _____________________________________________________________________________________________

A Survey on Classification and Rule Extraction Techniques for Datamining

“…But in Hybrid GA-based classifier [20], the performance is evaluated using many general data sets. There is an improvement in accuracy in [2], [3] and [7]. Compared to that, hybrid GA-based classifier model, it is having good improvement in classification accuracy.…”

“…Deborah R. Carvalho et al [2] proposed a hybrid decision tree/ GA method. In this hybrid approach, two specifically designed GA algorithms are used for discovering rules of examples belonging to small disjuncts and conventional decision tree algorithm are used for producing rules of examples belonging to large disjuncts.…”

Classification Rule Mining with Iterated Greedy

“…In this Section, we compare IG-RMiner with 10 other significant classification techniques, covering among others, evolutionary based techniques, ant colony optimization, fuzzy rule systems, and classic decision tree generators: ICRM [4], MPLCS [25], ILGA [26], CORE [27], SLAVE [28], GFS-GP [29], DTGA [30], AntMiner+ [31], RIPPER [32], C45R [33]. In particular, ICRM recently proved to be able to generate sufficiently accurate and easily interpretable rule classification systems.…”

A New Classification-Rule Pruning Procedure for an Ant Colony Algorithm

“…After an analysis of Ant Miner's original rule pruner, the following is a proposal to a new hybrid rule pruner, combining the original Ant-Miner's rule pruner with a rule pruner based on information gain -the latter somewhat inspired by the rule pruner proposed in [4]. (For a review of information gain in general, see [5].…”