In this paper, we propose an improved multi-label approach to classify web pages by genre. Our approach provides a multi-label classification scheme in which a web page can be assigned to more than one genre. To deal with the rapid evolution of web genres, our approach implements an incremental centroid-based classification scheme. Conducted experiments on a multi-labeled corpus of web pages show that our approach provides good results.
Published in:
Tools with Artificial Intelligence (ICTAI), 2011 23rd IEEE International Conference on
Date of Conference: 7-9 Nov. 2011