A Neural Network Pre-processor for a Fault Diagnosis Expert System
B67/210, B T LaboratoriesMartlesham Heath, Ipswich. IP5 7RE. U K.andyc@aom.bt.co.uk
, jamesb@aom.bt.co.uk
Abstract
Line test data is important for identifying and locating faults in access networks. However, the complexity of the data makes devising suitable evaluation heuristics difficult. This paper describes the use of a neural network, trained using historical data, both as a stand-alone system and as a pre-processor for a rule-based expert system. Comparisons between the different cases are drawn and recommendations are made for the applicability of the different approaches.
1 Introduction
In common with other telecommunication operators’ ...
Get Proceedings of the International Workshop on Applications of Neural Networks to Telecommunications now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.