Portal de Periódicos da CAPES

An ontology-supported database refurbishing technique and its application in mining actionable troubleshooting rules from real-life databases

2008; Elsevier BV; Volume: 21; Issue: 8 Linguagem: Inglês

10.1016/j.engappai.2008.04.015

1873-6769

Bong-Horng Chu, Cheng-En Lee, Cheng-Seen Ho,

Semantic Web and Ontologies

This article describes how database refurbishing and data mining technologies can be used to automatically construct a rule base for customer service systems that support troubleshooting in the global system for mobile communications (GSM) domain. With the support of the GSM troubleshooting ontology, a text-mining-empowered data refurbishing technique can successfully “refurbish” a real-life customer service database. The refurbished database then can support an association rule-mining algorithm that utilizes multiple minimum supports to discover better troubleshooting rules. We have developed an intelligent GSM troubleshooting system based on these techniques. Our experimental results with the system show a substantial improvement in both the quality and the quantity of the mined troubleshooting rules.