Memory-based classification with dynamic feature selection using self-organizing maps for pattern evaluation

Pateritsas, C; Stafylopatis, A

dc.contributor.author	Pateritsas, C	en
dc.contributor.author	Stafylopatis, A	en
dc.date.accessioned	2014-03-01T01:26:37Z
dc.date.available	2014-03-01T01:26:37Z
dc.date.issued	2007	en
dc.identifier.issn	0218-2130	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/18156
dc.subject	Classification	en
dc.subject	Feature weighting	en
dc.subject	Hybrid systems	en
dc.subject	Memory-based learning	en
dc.subject	Self-organizing maps	en
dc.subject.classification	Computer Science, Artificial Intelligence	en
dc.subject.classification	Computer Science, Interdisciplinary Applications	en
dc.subject.other	Classification (of information)	en
dc.subject.other	Learning algorithms	en
dc.subject.other	Learning systems	en
dc.subject.other	Mathematical models	en
dc.subject.other	Pattern recognition	en
dc.subject.other	Feature weighting	en
dc.subject.other	Hybrid systems	en
dc.subject.other	Memory-based learning	en
dc.subject.other	Self organizing maps	en
dc.title	Memory-based classification with dynamic feature selection using self-organizing maps for pattern evaluation	en
heal.type	journalArticle	en
heal.identifier.primary	10.1142/S0218213007003588	en
heal.identifier.secondary	http://dx.doi.org/10.1142/S0218213007003588	en
heal.language	English	en
heal.publicationDate	2007	en
heal.abstract	Memory-based learning is one of the main fields in the area of machine learning. We propose a new methodology for addressing the classification task that relies on the main idea of the k-nearest neighbors algorithm, which is the most important representative of this field. In the proposed approach, given an unclassified pattern, a set of neighboring patterns is found, not necessarily using all input feature dimensions. Also, following the concept of the naïve Bayesian classifier, we adopt the assumption of independence of input features in the outcome of the classification task. The two concepts are merged in an attempt to take advantage of their good performance features. In order to further improve the performance of our approach, we propose a novel weighting scheme of the memory-base. Using the self-organizing maps model during the execution of the algorithm, dynamic weights of the memory-base patterns are produced. Experimental results have shown improved performance of the proposed method in comparison with the aforementioned algorithms and their variations. © World Scientific Publishing Company.	en
heal.publisher	WORLD SCIENTIFIC PUBL CO PTE LTD	en
heal.journalName	International Journal on Artificial Intelligence Tools	en
dc.identifier.doi	10.1142/S0218213007003588	en
dc.identifier.isi	ISI:000251390300006	en
dc.identifier.volume	16	en
dc.identifier.issue	5	en
dc.identifier.spage	875	en
dc.identifier.epage	899	en