Rough Sets Methods Of Knowledge Acquisition In Incomplete Multi- Granular Information Systems

Rough set theory, which deals with incomplete information of data, was initiated by Pawlak, in 1982. The main idea of rough set theory is to obtain knowledge reduction by removing superfluous information to maintain classification ability and to induce concise decision rules from decision tables. Rough set data analysis uses only internal knowledge, avoids external parameters, and does not rely on prior model assumptions. With the development of the rough set theory, it has been successfully applied in many fields, especially, in the agricultural field in the over years.Granular computing is currently a new vivid direction in the research fields of artificial intelligence, especially, for intelligent information processing. It imitates human being's thinking and its objective is to establish effective computation models for dealing with large scale complex data and information. The purpose of granular computing is to seek for an approximation scheme which can effectively solve a complex problem at a certain level of granulation. Human beings often observe objects or deal with date hierarchically structured at different levels of granularities.The selection of a proper decision system in data set with hierarchical structures and the induction of corresponding decision rules at different levels of granulations are important issues in the study of intelligent systems.Aiming at knowledge acquisition in incomplete information systems with multi-granular labels, in this dissertation, a rough set approach to knowledge discovery in incomplete multi-scale decision systems from the perspective of granular computing is proposed. The concepts of incomplete information systems and rough set approximations with decision rules of incomplete decision systems are first reviewed. The notion of incomplete multi-scale information systems in the context of rough sets is then introduced. Information granules at different levels of scales in incomplete multi-scale information systems are then described. Lower and upper approximations with reference to different levels of scales in incomplete multi-scale information systems are also defined and their properties are examined. Optimal scaleselection with various requirements in incomplete multi-scale decision systems is further discussed. Relationships among different notions of optimal scales in incomplete multi-scale decision systems are presented. Finally, knowledge acquisitions in the sense of rule induction in consistent and inconsistent incomplete multi-scale decision systems are explored.