Data Resampling Techniques and Specific Algorithms Applied to a Critical Industrial Classification Problem

The paper deals with the problem of the detection of rare patterns in an unbalanced dataset related to an industrial problem concerning the identification of manufactured defective metal products on the basis of product and process parameters. Within this work several approaches have been attempted for the development of a classifier whose performance are able to meet the industrial requirements, i.e. a high rate of recognition of defective products. Considered the unbalanced nature of the available dataset, most known techniques used for dealing with this kind of databases (i.e. resampling techniques and specific algorithms) have been investigated and assessed, subsequently the most promising ones have been combined in order to exploit their advantages. This latter combination led to satisfactory results which make the developed classifier usable in the industrial field.