An Expert System for Financial Management Consultation.

概要

論文の詳細を見る
This paper describes our experiences with an expert system that integrates different sorts of expertise for financial management consultation and incorporates Constraint Satisfaction Paradigm (CSP) and Best-first Search (BFS) techniques to guide the consultation task to improve the performance of the inference engine in the previous work expert system. The task of financial consultation is achived based on the analysis of the enterprise's historical and financial information concerning its market characteristics, which determines a selection of management policies that are most beneficial to a customer, within an agreed set of standards. See figure 1.We begin the paper by specifying the overview of the expert system then present the Knowledge Structure and Inference Structure. In the former we describe the heuristics functions to avoid combinatorial explosion in the finding of feasible management policies. In the latter we describe the inference algorithm developed. We will handle a mechanism called the constraint solver, which propagate the financial index values in the financial consultation constraints.Finally we present conclusions and future work.
一般社団法人情報処理学会の論文
1993-09-27