作者: Anže Vavpetič , Vid Podpečan , Nada Lavrač
DOI: 10.1007/S10844-013-0292-1
关键词: Class (computer programming) 、 Semantic data mining 、 KEGG 、 Data mining 、 Gene expression profiling 、 Gene ontology 、 Natural language processing 、 Artificial intelligence 、 Computer science 、 Domain (software engineering) 、 Ontology (information science)
摘要: Subgroup discovery (SD) methods can be used to find interesting subsets of objects a given class. While subgroup describing rules are themselves good explanations the subgroups, domain ontologies provide additional descriptions data and alternative constructed rules. Such in terms higher level ontology concepts have potential providing new insights into investigation. We show that this explanatory power ensured by using recently developed semantic SD methods. present approach explaining subgroups through demonstrate its utility on motivational use case gene expression profiling where groups patients, identified expression, further explained from Gene Ontology KEGG orthology. qualitatively compare methodology with supporting factors technique for characterizing subgroups. The tools implemented within browser-based mining platform ClowdFlows.