Presentation | 2010-06-14 Hierarchical Pitman-Yor Topic Model Issei SATO, Hiroshi NAKAGAWA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Topic model is a probabilistic generative model that models latent semantics of words as "topic". In this paper, we propose a novel topic model that captures the power-law phenomenon of a word distribution, which is known as Zipf's law in linguistics. We use the Pitman-Yor process to model a generation process of a document. In an experiment using real data, our model outperformed LDA in document modeling in terms of perplexity. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Topic model / Pitman-Yor process |
Paper # | IBISML2010-7 |
Date of Issue |
Conference Information | |
Committee | IBISML |
---|---|
Conference Date | 2010/6/7(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 | Information-Based Induction Sciences and Machine Learning (IBISML) |
---|---|
Language | ENG |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Hierarchical Pitman-Yor Topic Model |
Sub Title (in English) | |
Keyword(1) | Topic model |
Keyword(2) | Pitman-Yor process |
1st Author's Name | Issei SATO |
1st Author's Affiliation | Graduate School of Information Sience and Technology, The University of Tokyo() |
2nd Author's Name | Hiroshi NAKAGAWA |
2nd Author's Affiliation | Interfaculty Initiative in Information Studies |
Date | 2010-06-14 |
Paper # | IBISML2010-7 |
Volume (vol) | vol.110 |
Number (no) | 76 |
Page | pp.pp.- |
#Pages | 7 |
Date of Issue |