Font Size: a A A

Random Information Systems And Knowledge Acquisition

Posted on:2006-07-07Degree:MasterType:Thesis
Country:ChinaCandidate:R H ChengFull Text:PDF
GTID:2208360185953730Subject:System theory
Abstract/Summary:PDF Full Text Request
Rough set theory, proposed by Pawlak Z. in the early 1980s, is a mathematical theory for reasoning about data. Since 1990s, it has received much attention of researchers and has become a flash point in the research area of computer science and information science. With more than twenty years development, rough set theory has been found to have very successful applications in the fields of artificial intelligence such as machine learning, pattern recognition, decision analysis, process control, knowledge discovery in databases, and expert system.The primitive notion in rough set theory is a pair of upper and lower approximation operators induced by the basic structure, an approximation space, consisting of a universe of discourse and a binary relation on it. The upper and lower approximation of a set characterize the non-numeric aspect of the setexpressed by the available information-the approximation space. In thisthesis, we make knowledge acquisition on rough set approximation operators and random set theory. The main results and originalities are summarized as follows:Some new concepts of knowledge reduction based on graded rough set theory such as upper(lower) approximation reduction and upper(lower) distribution reduction. The relationship among alternative reduction in inconsistent information systems are discussed.In random information systems and random dicision systems, according to the indiscernibility relation, the concept of indiscernibility attribute matrix is proposed. A reduction algorithm based on indiscernibility attribute matrix and evidence theory is introduced. Compared with discernibility matrix algorithm, from the example, this algorithm greatly reduces running time and memory space.
Keywords/Search Tags:rough sets, random sets, information systems, data mining, discernibility matrix, theory of evidence, mass function.
PDF Full Text Request
Related items