Presentation | 2010-05-28 Two-Tier Similarity Model in Story Link Detection Tadashi NOMOTO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The paper presents a novel approach to story link detection, whose goal is to determine whether a pair of news stories are linked, i.e., talk about the same event. The present work marks a departure from the prior work in that we measure similarity at two distinct levels of textual organization, namely, document and its cluster, and combine the scores to determine how well stories are linked. Experiments found that the present approach, which we call a 'two-tier similarity model,' comfortably beats the conventional approaches such as the cosine model and Clarity enhanced two-way KL divergence. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Story Link Detection / Topic Tracking / Similarity Measures / Relevance Feedback / Information Retrieval / Multi-lingual corpus / TDT |
Paper # | TL2010-3 |
Date of Issue |
Conference Information | |
Committee | TL |
---|---|
Conference Date | 2010/5/21(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Thought and Language (TL) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Two-Tier Similarity Model in Story Link Detection |
Sub Title (in English) | |
Keyword(1) | Story Link Detection |
Keyword(2) | Topic Tracking |
Keyword(3) | Similarity Measures |
Keyword(4) | Relevance Feedback |
Keyword(5) | Information Retrieval |
Keyword(6) | Multi-lingual corpus |
Keyword(7) | TDT |
1st Author's Name | Tadashi NOMOTO |
1st Author's Affiliation | National Institute of Japanese Literature() |
Date | 2010-05-28 |
Paper # | TL2010-3 |
Volume (vol) | vol.110 |
Number (no) | 63 |
Page | pp.pp.- |
#Pages | 5 |
Date of Issue |