%0 Journal Article %1 Wong:2000:JASIS %A Wong, Man Leung %A Leung, K. S. %A Cheng, K. S. %D 2000 %J Journal of the American Society for Information Science %K Computation, Data Evolutionary Learning Rule algorithms, genetic mining, programming, %P 870--881 %T Discovering Knowledge from Noisy Databases using Genetic Programming %U http://cptra.ln.edu.hk/~mlwong/journal/jasis2000.pdf %V 51 %X In data mining, we emphasise the need for learning from huge, incomplete and imperfect data sets (Fayyad et al. 1996, Frawley et al. 1991, Piatetsky-Shapiro and Frawley, 1991). To handle noise in the problem domain, existing learning systems avoid overfitting the imperfect training examples by excluding insignificant patterns. The problem is that these systems use a limiting attribute-value language for representing the training examples and the induced knowledge. Moreover, some important patterns are ignored because they are statistically insignificant. In this paper, we present a framework that combines Genetic Programming (Koza 1992; 1994) and Inductive Logic Programming (Muggleton, 1992) to induce knowledge represented in various knowledge representation formalisms from noisy databases. The framework is based on a formalism of logic grammars and it can specify the search space declaratively. An implementation of the framework, LOGENPRO (The Logic grammar based GENetic PROgramming system), has been developed. The performance of LOGENPRO is evaluated on the chess endgame domain. We compare LOGENPRO with FOIL and other learning systems in detail and find its performance is significantly better than that of the others. This result indicates that the Darwinian principle of natural selection is a plausible noise handling method which can avoid overfitting and identify important patterns at the same time. Moreover, the system is applied to one real-life medical database. The knowledge discovered provides insights to and allows better understanding of the medical domains.

@article{Wong:2000:JASIS, abstract = {In data mining, we emphasise the need for learning from huge, incomplete and imperfect data sets (Fayyad et al. 1996, Frawley et al. 1991, Piatetsky-Shapiro and Frawley, 1991). To handle noise in the problem domain, existing learning systems avoid overfitting the imperfect training examples by excluding insignificant patterns. The problem is that these systems use a limiting attribute-value language for representing the training examples and the induced knowledge. Moreover, some important patterns are ignored because they are statistically insignificant. In this paper, we present a framework that combines Genetic Programming (Koza 1992; 1994) and Inductive Logic Programming (Muggleton, 1992) to induce knowledge represented in various knowledge representation formalisms from noisy databases. The framework is based on a formalism of logic grammars and it can specify the search space declaratively. An implementation of the framework, LOGENPRO (The Logic grammar based GENetic PROgramming system), has been developed. The performance of LOGENPRO is evaluated on the chess endgame domain. We compare LOGENPRO with FOIL and other learning systems in detail and find its performance is significantly better than that of the others. This result indicates that the Darwinian principle of natural selection is a plausible noise handling method which can avoid overfitting and identify important patterns at the same time. Moreover, the system is applied to one real-life medical database. The knowledge discovered provides insights to and allows better understanding of the medical domains.}, added-at = {2008-06-19T17:35:00.000+0200}, author = {Wong, Man Leung and Leung, K. S. and Cheng, K. S.}, biburl = {https://www.bibsonomy.org/bibtex/2e621b619bd75cb7e9d2795a92d0f9b2a/brazovayeye}, interhash = {1c9b3f2e29890ecd8e7727974ca5f1e3}, intrahash = {e621b619bd75cb7e9d2795a92d0f9b2a}, journal = {Journal of the American Society for Information Science}, keywords = {Computation, Data Evolutionary Learning Rule algorithms, genetic mining, programming,}, pages = {870--881}, timestamp = {2008-06-19T17:54:23.000+0200}, title = {Discovering Knowledge from Noisy Databases using Genetic Programming}, url = {http://cptra.ln.edu.hk/~mlwong/journal/jasis2000.pdf}, volume = 51, year = 2000 }