Multi-Label Wikipedia Classification with Textual and Link Features

Multi-Label Wikipedia Classification with Textual and Link Features - scientific work related to Wikipedia quality published in 2009, written by Boris Chidlovskii.

Overview

Authors address the problem of categorizing a large set of linked documents with important content and structure aspects, in particular, from the Wikipedia collection proposed at the INEX 2009 XML Mining challenge. Authors analyze the network of collection pages and turn it into valuable features for the classification. Authors combine the content-based and link-based features of pages to train an accurate categorizer for unlabelled pages. In the multi-label setting, authors revise a number of existing techniques and test some which show a good scalability. Authors report evaluation results obtained with a variety of learning methods and techniques on the training set of the Wikipedia corpus.