Parameter tuning for the MAX expert system

Author

Merz, C.J. ; Pazzani, Michael

Author_Institution

Dept. of Inf. & Comput. Sci., California Univ., Irvine, CA, USA

fYear

1994

fDate

6-9 Nov 1994

Firstpage

632

Lastpage

639

Abstract

We investigate methods for tuning numeric parameters in Nynex MAX, a telephone trouble screening expert system. Steepest descent, hillclimbing, and simulated annealing parameter adjustment strategies are applied to the problems of maximizing classification accuracy and minimizing misclassification cost. For both of those optimization problems we evaluate each algorithm´s ability to tune initial parameters for several situations

Keywords

diagnostic expert systems; simulated annealing; telecommunication computing; telephony; MAX expert system; Nynex MAX; Steepest descent; classification accuracy; hillclimbing; initial parameters; misclassification cost; numeric parameters; optimization problems; parameter tuning; simulated annealing parameter adjustment strategies; telephone trouble screening expert system; Algorithm design and analysis; Central office; Cost function; Error analysis; Expert systems; Simulated annealing; Telephony; Testing; Voltage; Wiring;

fLanguage

English

Publisher

ieee

Conference_Titel

Tools with Artificial Intelligence, 1994. Proceedings., Sixth International Conference on

Conference_Location

New Orleans, LA

Print_ISBN

0-8186-6785-0

Type

conf

DOI

10.1109/TAI.1994.346450

Filename

346450

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2302426