%0 Conference Paper %1 Lavangnananda_2004_CIMCA %A Lavangnananda, K. %B 2004 International Conference on Computational Intelligence for Modelling, Control and Automation - Cimca'2004 %C Gold Coast, Australia %D 2004 %E Mohammadian, Masoud %K algorithms, computation, data evolutionary genetic inductive learning mining, programming, %P 279--290 %T A Genetic Programming Approach to Inductive Learning %U http://dummy/Lavangnananda_2004_CIMCA.pdf %X There have been many applications of artificial intelligence data mining recently. One of its many benefits includes the ability to cluster or generate patterns from large amount of data when conventional statistical methods are proven ineffective. One such techniques in data mining is inductive learning. There have been applications of evolutionary computation in inductive learning where genetic algorithms have been employed in chromosomes representation. This paper describes an attempt to use genetic programming in inductive learning. A program known as Genetic Programming for Inductive Learning (GPIL) is described. It uses genetic programming and rectifies the short comings of chromosomes representation in genetic algorithms. The program has been tested on a benchmark data set. It achieved better performance with higher accuracy than previous works on the same data set. The paper also discusses relevant aspects in using genetic programming in inductive learning and suggests directions for future work.

@inproceedings{Lavangnananda_2004_CIMCA, abstract = {There have been many applications of artificial intelligence data mining recently. One of its many benefits includes the ability to cluster or generate patterns from large amount of data when conventional statistical methods are proven ineffective. One such techniques in data mining is inductive learning. There have been applications of evolutionary computation in inductive learning where genetic algorithms have been employed in chromosomes representation. This paper describes an attempt to use genetic programming in inductive learning. A program known as Genetic Programming for Inductive Learning (GPIL) is described. It uses genetic programming and rectifies the short comings of chromosomes representation in genetic algorithms. The program has been tested on a benchmark data set. It achieved better performance with higher accuracy than previous works on the same data set. The paper also discusses relevant aspects in using genetic programming in inductive learning and suggests directions for future work.}, added-at = {2008-06-19T17:35:00.000+0200}, address = {Gold Coast, Australia}, author = {Lavangnananda, K.}, biburl = {https://www.bibsonomy.org/bibtex/2203e7e6228d424c33e866b74e63d7973/brazovayeye}, booktitle = {2004 International Conference on Computational Intelligence for Modelling, Control and Automation - Cimca'2004}, editor = {Mohammadian, Masoud}, interhash = {8eede60441ed1be6cfe274b65d09a90f}, intrahash = {203e7e6228d424c33e866b74e63d7973}, keywords = {algorithms, computation, data evolutionary genetic inductive learning mining, programming,}, month = {12-14 July}, notes = {http://www.ise.canberra.edu.au/ An early version of \cite{Lavangnananda:2006:ieeeMWALS}}, pages = {279--290}, size = {12 pages}, timestamp = {2008-06-19T17:45:12.000+0200}, title = {A Genetic Programming Approach to Inductive Learning}, url = {http://dummy/Lavangnananda_2004_CIMCA.pdf}, year = 2004 }