الفهرس 
CHAPTER 1 : Introduction to Decision MakingOnly 14 pages are availabe for public view
 |  | from | 172 |  |  |
| | | from | 172 | | |
Abstract
Decision making is key in every part of a person’s life; everyone makes decisions from the moment they wake up in the morning until they go to bed at night. These range from the simple decisions to the more complex. This thesis deals with the problem of knowledge acquisition for decision-support systems in order to improve the decision-making process of experts. In most real-world problems it is necessary to obtain explicit knowledge by which the right decisions can be made and explained. The knowledge is acquired by learning a set of rules from examples. It is argued that the result of learning should be a rule base that fits both the actual data (data fit) and the user’s frame of reference (mental fit).
Rule induction is the extraction of knowledge from data. If data is thought of as raw facts and figures of limited direct use in decision-making, then knowledge may be defined as a relevant and useable summary of that data, a necessity to the operational, managerial and strategic planning processes of an organization Another important use of rule induction is within the context of developing intelligent reasoning systems. High-level knowledge, usually in the form of “If ……. Then” rules, is required for the decision-making processes of such systems. Traditionally this has been obtained via discussions with domain experts, though this approach has many problems and shortcomings. The interviews are generally long, inefficient and frustrating for both the domain experts and knowledge engineers, especially so in domains where experts make decisions based on incomplete or imprecise information. This knowledge acquisition phase is often the main bottleneck within the knowledge engineering process and therefore considerable effort has been expended in designing rule induction algorithms that automatically or semi-automatically generate useful knowledge from detailed historical data already available.
In this thesis a few of the most common approaches to rule induction are outlined below, with their relative merits and limitations. Another main contribution of this thesis is new rule induction algorithms: the first based on Symbolic Value Partition Technique and Generalized Distribution Table, another one is based on Fuzzy Logic and Inclusion Degree Theory, the last one based on Rough set and Decision Network. Finally, Theses algorithms are applied to real Engineering application such as Fault Diagnosis of Electrical Distribution System, an On-line Cable Condition Monitoring System, and Turbine-Generator Unit Fault Diagnosis This thesis consists of six main chapters; these chapters can be described in the following manner:
CHAPTER 1 : In this chapter Most of the background information and definitions of Decision making necessary to understand the thesis has been placed, and it is organized as follows : section
1.1 give an introduction, In section
1.2 some Concepts and Definitions are introduced, In sections 1.3 several Kinds of Decisions are illustrated. In section 1.4 Decision Making Procedure was explained, In sections 1.5 and 1.6 Approaches to Decision Making Some Decision Making Strategies are presented. In section 1.7 give short notes on Multi-attribute decision making methods.
CHAPTER 2 : This chapter gives an introductory to multicriteria decision analysis (MCDA).it also gives an introductory description of rough set theory , how it used to decision support and show that The rough set approach is intended to deal with inconsistency in multicriteria decision analysis (MCDA). This chapter is organized as follows: In Section 2.2, a ٢-general view of the rough set approach is given. In Section 2.3, we introduce a distinction between classification and sorting problems.
CHAPTER 3 : This chapter Introduce rule–generation algorithm of literature review. It is organized as follows: In Section 3.1 a few of the most common approaches to rule induction are outlined, with their relative merits and limitations. Section 3.4.1 discusses the C4.5 algorithm, which is the improved version of ID3 and Section 3.4.2 discusses the CN2 algorithm which is an improvement over AQ15.
CHAPTER 4 : In this chapter three new rule induction algorithms are proposed and applied to examples found in the literature review as well as newly constructed. It is organized as follows: In Section 4.1 some Preliminaries are illustrated, the main idea of the first algorithm which based on Symbolic Value Partition Technique and Generalized Distribution Table was explained, then the OSGP-problem was proposed as the basis of the algorithm and finally list the algorithm. In Section 4.2 some Preliminaries are illustrated, the main idea of the second algorithm which based on Fuzzy Logic and Inclusion Degree Theory, an example is given to show how the proposed method can be used to generate diagnostic rules. In Section 4.3 some Preliminaries are illustrated, the main idea of the third algorithm which based on Fuzzy Logic and Inclusion Degree Theory, an example is given to show how the proposed method can be used to generate decision rules.
CHAPTER 5 : In this chapter the proposed algorithms are applied to real Engineering application such as Fault Diagnosis of Electrical Distribution System, an On-line Cable Condition Monitoring System, and Turbine-Generator Unit Fault Diagnosis. And test results are presented.
CHAPTER 6 : This chapter describes some concluding remarks, recommendations and some points for future researches.