LDTM: A Latent Document Type Model for Cumulative Citation Recommendation

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 561–566, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. LDTM: A Latent Document Type Model for Cumulative Citation Recommendation Jingang Wang1∗, Dandan Song1†, Zhiwei Zhang2, Lejian Liao1, Luo Si2, Chin-Yew Lin3 1School of Computer Science, Beijing Institute of Technology 2Dept. of Computer Science, Purdue University 3Knowledge Mining Group, Microsoft Research {bitwjg, sdd, liaolj}@bit.edu.cn {zhan1187, lsi}@purdue.edu cyl@microsoft.com Abstract This paper studies Cumulative Citation Recommendation (CCR) - given an entity in Knowledge Bases, how to effectively detect its potential citations from volume text streams. Most previous approaches treated all kinds of features indifferently to build a global relevance model, in which the prior knowledge embedded in docu- ments cannot be exploited adequately. To address this problem, we propose a la- tent document type discriminative model by introducing a latent layer to capture the correlations between documents and their underlying types. The model can better adjust to different types of documents and yield flexible performance when dealing with a broad range of document types. An extensive set of experiments has been con- ducted on TREC-KBA-2013 dataset, and the results demonstrate that this model can yield a significant performance gain in rec- ommendation quality as compared to the state-of-the-art. 1 Introduction Knowledge Bases (KBs), like Wikipedia, are playing increasingly important roles in numerous entity-based information retrieval tasks. Neverthe- less, most KBs are hard to be up-to-date due to their manual maintenances by human editors. As reported in (Frank et al., 2012), there exists a me- dian time lag of 356 days between the day a news article is published and the time that the news is cited in a Wikipedia article dedicated to the entity concerned by the news. The time lag would be reduced if relevant documents could be automati- cally detected as soon as they are published online ∗This work was partially performed when the first author was visiting Purdue University and Microsoft Research Asia. †Corresponding Author and then recommended to the editors. This task is studied as Cumulative Citation Recommendation (CCR). Formally, given a set of KB entities, CCR is to filter relevant documents from a stream cor- pus and evaluate their citation-worthiness to the target entities. A variety of supervised approaches (e.g., clas- sification, learning to rank) have been employed and achieved promising results (Wang et al., 2013; Balog and Ramampiaro, 2013; Balog et al., 2013). Nevertheless, most of them leverage all features indiscriminately to build a global relevance model, which leads to unsatisfactory performance. The documents can offer some prior knowledge, which is named as type in this paper. The type is the prior knowledge embedded in the document that im- pacts on the probability of its being recommended to KBs. For instance, when dealing with a docu- ment on “music” topic, we would like to have less weights put on a politician entity because this doc- ument is not likely to related to it, but more often related to musicians or musical bands. …

Content truncated to 500 words