论文信息 - Topic Tracking Based on Keywords Dependency Profile

Topic Tracking Based on Keywords Dependency Profile

Topic tracking is an important task of Topic Detection and Tracking (TDT). Its purpose is to detect stories, from a stream of news, related to known topics. Each topic is "known" by its association with several sample stories that discuss it. In this paper, we propose a new method to build the keywords dependency profile (KDP) of each story and track topic basing on similarity between the profiles of topic and story. In this method, keywords of a story are selected by document summarization technology. The KDP is built by keywords co-occurrence frequency in the same sentences of the story. We demonstrate this profile can describe the core events in a story accurately. Experiments on the mandarin resource of TDT4 and TDT5 show topic tracking system basing on KDP improves the performance by 13.25% on training dataset and 7.49% on testing dataset comparing to baseline.

Wei Zheng | Yu Hong | Ting Liu | Yu Zhang | Jili Fan

[1] David L. Waltz,et al. Classifying news stories using memory based reasoning , 1992, SIGIR '92.

[2] Victor Lavrenko,et al. Language-specific models in multilingual topic tracking , 2004, SIGIR '04.

[3] James Allan,et al. Topic detection and tracking: event-based information organization , 2002 .

[4] Dragomir R. Radev,et al. Centroid-based summarization of multiple documents: sentence extraction, utility-based evaluation, and user studies , 2000, ArXiv.

[5] Qin Lu,et al. Topic tracking with time granularity reasoning , 2006, TALIP.

[6] Juha Makkonen,et al. Investigations on Event Evolution on TDT , 2003, NAACL.

[7] James Allan,et al. On-Line New Event Detection and Tracking , 1998, SIGIR.

[8] Yiming Yang,et al. CMU Report on TDT-2: Segmentation, Detection and Tracking , 1999 .

[9] Chirag Shah,et al. Representing documents with named entities for story link detection (SLD) , 2006, CIKM '06.

[10] James Allan,et al. Topic Detection and Tracking , 2002, The Information Retrieval Series.

[11] Qin Lu,et al. Profile-based event tracking , 2005, SIGIR '05.

[12] Yiming Yang,et al. Topic-conditioned novelty detection , 2002, KDD.

[13] Francine Chen,et al. A trainable document summarizer , 1995, SIGIR '95.