Interestingness Measures In Rule Mining: A Valuation
Garima Garima, Jyoti Vashishtha · International Journal of Engineering Research and Applications · 2014
In data mining it is normally desirable that discovered knowledge should possess characteristics such as accuracy, comprehensibility and interestingness. The vast majority of data mining algorithms generate patterns that are accurate and reliable but they might not be interesting. Interestingness measures are used to find the truly interesting rules which will help the user in decision making in exceptional state of affairs. A variety of interestingness measures for rule mining have been suggested by researchers in the field of data mining. In this paper we are going to carry out a valuation of these interestingness measures and classify them from numerous perspectives.