Competition-based induction of decision models from examples

Abstract

Abstract. Symbolic induction is a promising approach to constructing decision models by extracting regularities from a data set of examples. The predominant type of model is a classification rule (or set of rules) that maps a set of relevant environmental features into specific categories or values. Classifying loan risk based on borrower profiles, consumer choice from purchase data, or supply levels based on operating conditions are all examples of this type of model-building task. Although current inductive approaches, such as ID3 and CN2, perform well on certain problems, their potential is limited by the incremental nature of their search. Genetic ,algorithms (GA) have shown great promise on complex search domains, and hence suggest a means for overcoming these limitations. However, effective use of genetic search in this context requires a framework that promotes the fundamental model-building objectives of predictive accuracy and model simplicity. In this article we describe, COGIN, a GAbased inductive system that exploits the conventions of induction from examples to provide this framework. The novelty of COGIN lies in its use of training set coverage to simultaneously promote competition in various classification niches within the model and constrain overall model complexity. Experimental comparisons 'with NewID and CN2 provide evidence of the effectiveness of the COGIN framework and the viability of the GA approach.

Genetic fuzzy systems: taxonomy, current research trends and prospects

“…In this model the chromosomes compete and cooperate simultaneously. COGIN [49], REGAL [44] and LOGENPRO [106] are examples with this kind of representation.…”

Data-oriented analyses of ciliate foraging behaviors

“…The basic concept of decision tree induction is based on a greedy algorithm that constructs a decision tree in a top-down, recursive, and divideand-conquer manner (Greene & Smith, 1993). The ID3 algorithm (Quinlan, 1986) is an iterative decision tree induction algorithm that generates simple trees with higher predictive accuracy instead of complex trees.…”

REGAL-TC: a distributed genetic algorithm for concept learning based on REGAL and the treatment of counterexamples

“…In this model, the chromosomes compete and cooperate simultaneously. COGIN (Greene and Smith 1993), REGAL (Giordana and Neri 1995), G-Net (Giordana et al 1997), OCEC (Jiao et al 2006), EDGAR (Rodríguez et al 2010) and DOGMA (Hekanaho 1997) are examples that can be located in this framework.…”

“…These are OCEC (Jiao et al 2006) and COGIN (Greene and Smith 1993) as GCCL methods. Due to CORE (Tan et al 2006b) has shown a low performance, we decided to replace it by GIL (Janikow 1993), which is a GBML algorithm that only works with nominal values, as in our proposal.…”