We propose a novel attention network for document annotation with user-generated tags. The network is designed according to the human reading and annotation behaviour. Usually, users try to digest the title and obtain a rough idea about the topic first, and then read the content of the document. Present research shows that the title metadata could largely affect the social annotation. To better utilise this information, we design a framework that separates the title from the content of a document and apply a title-guided attention mechanism over each sentence in the content. We also propose two semanticbased loss regularisers that enforce the output of the network to conform to label semantics, i.e. similarity and subsumption. We analyse each part of the proposed system with two real-world open datasets on publication and question annotation. The integrated approach, Joint Multi-label Attention Network (JMAN), significantly outperformed the Bidirectional Gated Recurrent Unit (Bi-GRU) by around 13%-26% and the Hierarchical Attention Network (HAN) by around 4%-12% on both datasets, with around 10%-30% reduction of training time.
R. Jäschke, L. Marinho, A. Hotho, L. Schmidt-Thieme, und G. Stumme. Knowledge Discovery in Databases: PKDD 2007, 11th European Conference on Principles and Practice of Knowledge Discovery in Databases, Warsaw, Poland, September 17-21, 2007, Proceedings, Volume 4702 von Lecture Notes in Computer Science, Seite 506-514. Springer, (2007)
C. Veres. Natural Language Processing and Information Systems, Volume 3999 von Lecture Notes in Computer Science, Seite 58--69. Berlin/Heidelberg, Springer, (2006)
F. Suchanek, M. Vojnovic, und D. Gunawardena. CIKM '08: Proceeding of the 17th ACM conference on Information and knowledge management, Seite 223--232. New York, NY, USA, ACM, (2008)
J. Illig, A. Hotho, R. Jäschke, und G. Stumme. Knowledge Processing and Data Analysis, Volume 6581 von Lecture Notes in Computer Science, Seite 136--149. Berlin/Heidelberg, Springer, (2011)
A. Shepitsen, J. Gemmell, B. Mobasher, und R. Burke. RecSys '08: Proceedings of the 2008 ACM conference on Recommender systems, Seite 259--266. New York, NY, USA, ACM, (2008)
P. Heymann, D. Ramage, und H. Garcia-Molina. SIGIR '08: Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Seite 531--538. New York, NY, USA, ACM, (2008)
C. Cattuto, D. Benz, A. Hotho, und G. Stumme. The Semantic Web - ISWC 2008, Proc.Intl. Semantic Web Conference 2008, Volume 5318 von LNAI, Seite 615--631. Heidelberg, Springer, (2008)
P. Symeonidis, A. Nanopoulos, und Y. Manolopoulos. RecSys '08: Proceedings of the 2008 ACM conference on Recommender systems, Seite 43--50. New York, NY, USA, ACM, (2008)