dc.contributor.author | Karatsiolis, Savvas | en |
dc.contributor.author | Schizas, Christos N. | en |
dc.contributor.editor | Gammerman A. | en |
dc.contributor.editor | Vovk V. | en |
dc.contributor.editor | Papadopoulos H. | en |
dc.creator | Karatsiolis, Savvas | en |
dc.creator | Schizas, Christos N. | en |
dc.date.accessioned | 2019-11-13T10:40:38Z | |
dc.date.available | 2019-11-13T10:40:38Z | |
dc.date.issued | 2015 | |
dc.identifier.issn | 0302-9743 | |
dc.identifier.uri | http://gnosis.library.ucy.ac.cy/handle/7/54224 | |
dc.description.abstract | We discuss a way of implementing feature mapping for classification problems by expressing the given data through a set of functions comprising of a mixture of convex functions. In this way, a certain pattern’s potential of belonging to a certain class is mapped in a way that promotes interclass separation, data visualization and understanding of the problem’s mechanics. In terms of enhancing separation, the algorithm can be used in two ways: to construct problem features to feed a classification algorithm or to detect a subset of problem attributes that could be safely ignored. In terms of problem understanding, the algorithm can be used for constructing a low dimensional feature mapping in order to make problem visualization possible. The whole approach is based on the derivation of an optimization objective which is solved with a genetic algorithm. The algorithm was tested under various datasets and it is successful in providing improved evaluation results. Specifically for Wisconsin breast cancer problem, the algorithm has a generalization success rate of 98% while for Pima Indian diabetes it provides a generalization success rate of 82%. © Springer International Publishing Switzerland 2015. | en |
dc.source | 3rd International Symposium on Statistical Learning and Data Sciences, SLDS 2015 | en |
dc.source.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-84949816060&doi=10.1007%2f978-3-319-17091-6_3&partnerID=40&md5=6c2d4148e1aef47dbd8fb0bef7e7f545 | |
dc.subject | Visualization | en |
dc.subject | Optimization | en |
dc.subject | Algorithms | en |
dc.subject | Genetic algorithms | en |
dc.subject | Functions | en |
dc.subject | Mapping | en |
dc.subject | Classification | en |
dc.subject | Classification (of information) | en |
dc.subject | Convex functions | en |
dc.subject | Evaluation results | en |
dc.subject | Data visualization | en |
dc.subject | Classification algorithm | en |
dc.subject | Convexity | en |
dc.subject | Features | en |
dc.subject | Flow visualization | en |
dc.subject | Inter-class distance | en |
dc.subject | Pima Indian Diabetes | en |
dc.subject | Problem understanding | en |
dc.title | Feature mapping through maximization of the atomic interclass distances | en |
dc.type | info:eu-repo/semantics/article | |
dc.identifier.doi | 10.1007/978-3-319-17091-6_3 | |
dc.description.volume | 9047 | |
dc.description.startingpage | 75 | |
dc.description.endingpage | 85 | |
dc.author.faculty | 002 Σχολή Θετικών και Εφαρμοσμένων Επιστημών / Faculty of Pure and Applied Sciences | |
dc.author.department | Τμήμα Πληροφορικής / Department of Computer Science | |
dc.type.uhtype | Article | en |
dc.description.notes | <p>Sponsors: British Classification Society | en |
dc.description.notes | British Computer Society | en |
dc.description.notes | et al | en |
dc.description.notes | Paris Dauphine University | en |
dc.description.notes | Royal Statistical Society | en |
dc.description.notes | University of London | en |
dc.description.notes | Conference code: 158849</p> | en |
dc.source.abbreviation | Lect. Notes Comput. Sci. | en |
dc.contributor.orcid | Schizas, Christos N. [0000-0001-6548-4980] | |
dc.gnosis.orcid | 0000-0001-6548-4980 | |