A Genetic Algorithm-Based Approach to Data Mining

Most data mining systems to date have used variants of traditional machine-learning algorithms to tackle the task of directed knowledge discovery. This paper presents an approach which, as well as being useful for such directed data mining, can also be applied to the further tasks of undirected data mining and hypothesis refinement. This approach exploits parallel genetic algorithms as the search mechanism and seeks to evolve explicit "rules" for maximum comprehensibility. Example rules found in real commercial datasets are presented.

[1]  John J. Grefenstette,et al.  A Parallel Genetic Algorithm , 1987, ICGA.

[2]  Filippo Neri,et al.  Search-Intensive Concept Induction , 1995, Evolutionary Computation.

[3]  J. E. Gibson,et al.  Adaptive Learning Systems , 2017 .

[4]  Gregory Piatetsky-Shapiro,et al.  Discovery, Analysis, and Presentation of Strong Rules , 1991, Knowledge Discovery in Databases.

[5]  Gilles Venturini,et al.  Learning First Order Logic Rules with a Genetic Algorithm , 1995, KDD.

[6]  Padhraic Smyth,et al.  Rule Induction Using Information Theory , 1991, Knowledge Discovery in Databases.

[7]  Patrick D. Surry,et al.  RPL2: A Language and Parallel Framework for Evolutionary Computing , 1994, PPSN.

[8]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[9]  William J. Frawley,et al.  Using Functions to Encode Domain and Contextual Knowledge in Statistical Induction , 1991, Knowledge Discovery in Databases.

[10]  Stephen F. Smith,et al.  A learning system based on genetic adaptive algorithms , 1980 .

[11]  Willi Klösgen,et al.  Exploration of Simulation Experiments by Discovery , 1994, KDD Workshop.

[12]  Jan M. Zytkow,et al.  Interactive Mining of Regularities in Databases , 1991, Knowledge Discovery in Databases.

[13]  John H. Holland,et al.  Escaping brittleness: the possibilities of general-purpose learning algorithms applied to parallel rule-based systems , 1995 .

[14]  Stephen F. Smith,et al.  Using Coverage as a Model Building Constraint in Learning Classifier Systems , 1994, Evolutionary Computation.

[15]  N. J. Radcliffe,et al.  GA-MINER: Parallel Data Mining with Hierarchical Genetic Algorithms Final Report , 1995 .

[16]  Filippo Neri,et al.  A Parallel Genetic Algorithm for Concept Learning , 1995, ICGA.