Categories and Subject Descriptors

This paper studies how to reduce the amount of human su-pervision for identifying splogs / authentic blogs in the con-text of continuously updating splog data sets year by year. Following the previous works on active learning, against the task of splog / authentic blog detection, this paper empir-ically examines several strategies for selective sampling in active learning by Support Vector Machines (SVMs). As a confidence measure of SVMs learning, we employ the dis-tance from the separating hyperplane to each test instance, which have been well studied in active learning for text clas-sification. Unlike those results of applying active learning to text classification tasks, in the task of splog / authentic blog detection of this paper, it is not the case that adding least confident samples peforms best