Chapter V

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 
136 137 138 139 140 141 142 143 144 145 146 

Belief Function

Approach to

Evidential Reasoning

in Causal Maps

Rajendra P. Srivastava

University of Kansas, USA

Mari W. Buche

Michigan Technological University, USA

Tom L. Roberts

University of Kansas, USA

Abstract

The purpose of this chapter is to demonstrate the use of the evidential reasoning

approach under the Dempster-Shafer (D-S) theory of belief functions to analyze

revealed causal maps (RCM). The participants from information technology (IT)

organizations provided the concepts to describe the target phenomenon of Job

Satisfaction. They also identified the associations between the concepts. This chapter

discusses the steps necessary to transform a causal map into an evidential diagram. The

evidential diagram can then be analyzed using belief functions technique with survey

data, thereby extending the research from a discovery and explanation stage to testing

and prediction. An example is provided to demonstrate these steps. This chapter also

provides the basics of Dempster-Shafer theory of belief functions and a step-by-step

description of the propagation process of beliefs in tree-like evidential diagrams.