Data mining applications in BT

Abstract

Presents a number of case studies demonstrating how data mining is being used within BT to discover valuable knowledge. The case studies highlight the use of these techniques, the wealth of information contained in databases and learning points encountered. The studies presented are in the following areas: (1) identifying faults on printed circuit boards; (2) discovering the organisational structure of groups of criminals; and (3) predicting outcomes of credit assessment and litigation. Data mining techniques have been shown to provide significant benefits in terms of cost savings and detection of fraud against the company. What has also been shown, particularly in the case of fraud, is that the combination of data mining, data visualisation and human expertise is highly effective. A number of lessons have been learnt from these studies. First, simply throwing a machine learning system at a database is unlikely to yield good results. A significant amount of effort is required to pre-process data and understand its meaning in the problem domain Specialist domain knowledge will almost certainly be required. Second, a good deal of problem simplification is likely to be needed if high-accuracy results are to be obtained. This inevitably requires an element of compromise between overall business goals and what is practically achievable. Lastly, data mining alone will not yield business benefits. To be successful, it is necessary that business processes are changed to deliver them. The mind-set which views data as something to be archived has to be changed to one which views it as a valuable resource.