Affiliations: [a]
Key Laboratory of Complex System Optimization and Big Data Processing in Department of Guangxi Education, Yulin Normal University, Yulin, Guangxi, P.R. China | [b]
School of Science, Guangxi University for Nationalities, Nanning, Guangxi, P.R. China | [c]
Key Laboratory of Complex System Optimization and Big Data Processing in Department of Guangxi Education, Yulin Normal University, Yulin, Guangxi, P.R. China
Correspondence:
[*]
Corresponding author. Pei Wang, Key Laboratory of Complex System Optimization and Big Data Processing in Department of Guangxi Education, Yulin Normal University, Yulin, Guangxi 537000, P.R. China. E-mail: [email protected].
Abstract: An information system as a database that represents relationships between objects and attributes is an important model in the field of artificial intelligence. A three-source information system of this paper is an information system where there exist three data: categorical, boolean and set-valued data. This paper explores uncertainty measurement for this kind of information system. The concept of a three-source information system is first described by means of set matrices. Then, information structures in a three-source information system are presented and relationships between information structures are studied from the two aspects of dependence and separation. Next, properties of information structures in a three-source information system are given by using inclusion degree. Finally, as an application for information structures, uncertainty measurement for a three-source information system are investigated by means of its information structures. These results will be helpful for understanding the essence of uncertainty in a three-source information system.
Keywords: Three-source information system, information structure, dependence, inclusion degree, entropy, uncertainty, measurement
