Learning ontology aware classifiers

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" Learning ontology aware classifiers "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	149222
Doc. No	:	ET21014
Main Entry	:	Jun Zhang
Title Proper	:	Learning ontology aware classifiers
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Many applications of data-driven knowledge discovery processes call for the exploration ofdata from multiple points of view that reflect different ontological commitments on the part ofthe learner. This is particularly important in scientific applications in which specific ontologicaland representational commitments often reflect prior knowledge and working assumptions. Ofparticular interest in this context are algorithms for learning classifiers from ontologies anddata. In many practical applications, it is often the case that the instances to be classified arespecified using attributes at different level of precision, leading to partially specified instances.Against this background, my dissertation research is aimed at the design and analysis ofalgorithms for the construction of robust, compact, accurate and ontology-aware classifiers.We have precisely formulated the problem of learning pattern classifiers from AttributeValue Taxonomies (i.e., AVT, a specific class of ontologies) and partially specified data. Wehave designed and implemented efficient algorithms for learning classifiers from such data.
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL44161.pdf
Title and statement of responsibility and	:	Learning ontology aware classifiers [Thesis]