ABEL CLASSIFICATION FRAMEWORK BASED ON VPRS 6 . 1 . 2 Compound Text Classification Framework

With the diversification of the TC task, to construct multi-label classifier is often more in line with the needs of practical applications. However, in a multi-label classification task, each document often corresponds to more than one class label. In this chapter, we will construct a compound classification framework which may transform a multi-label classification task into several single label classifications and several simplified multi-label classifications. Rough Set (RS), as a mathematical tool dealing with vague and uncertain knowledge, can be appropriately applied in depicting data distribution and calculating extent of the region. This chapter presents a compound multi-label classification framework based on Variable Precision Rough Set (VPRS), which partitions the document sample space, thus the multi-label classification task is decomposed into a number of simplified classification problems.