|
Title:
|
UNCERTAINTY MODELING IN EXPERT SYSTEMS FOR FINANCIAL RELIABILITY EVALUATION1 |
|
Author(s):
|
Bosnjak Zita , Bosnjak Sasa |
|
ISBN:
|
972-98947-0-1 |
|
Editors:
|
António Palma dos Reis and Pedro Isaías |
|
Year:
|
2003 |
|
Edition:
|
1 |
|
Keywords:
|
Fuzzy expert systems, LOWA operators, linguistic quantifiers. |
|
Type:
|
Full Paper |
|
First Page:
|
452 |
|
Last Page:
|
459 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
In this paper we described an expert system for financial reliability evaluation of companies, from the perspective of the improvements we have made over time in the management of uncertainty and imprecision inherent to data relevant for the problem solution. The first version of this expert system was a crisp one - it was based on the decision making model that utilized a hierarchical aggregation structure, where the financial reliability estimation was positioned on the top of a hierarchy and the leaves represented the attributes that had impact on the evaluation. The later version of the expert system takes advantage from the inclusion of fuzzy set theory into the financial reliability assessment. The fuzzy expert system is based on fuzzified decision model in which the subjective estimations of experts are expressed as linguistic terms, by fuzzy sets. The rational behind this model is described by production rules. For the aggregation of the assessments we use the combination of the LOWA operator, the fuzzy linguistic quantifier most and the classical min-max operator, while the center-of-gravity method is used as a defuzzification method, as we require the best compromise among the evaluated degrees of financial reliability. The results and experiences of the transformation of the crisp system into a fuzzy application are presented in this paper. |
|
|
|
|
|
|
