Presentation | 2007-05-19 Keywords analysis of electronic bulletin board using the multivariate latent logistic model Maomi UENO, Toshio OKAMOTO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This study proposes (1) A new definition of the kewords weights in the morphological analysis of the discussion in the electro bullitein board and it's application to the automatic keywords extraction system, and (2) Visualization method of the keywords characteristics and the contributers characteristics using the multivariate latent logistic model. The advantages of the proposed method are 1. the estimators have the asymptotic consistency using the maximum likelihood method or Bayesian estimation method, 2. The method can decide the optimum number of the parameters using the information criteria as like MDL, 3. It is easy to define the keywords weights, they can be applied to the keywords analysis and abstraction system, 4. the multivariate latent logistic model is easy to calculate the huge data, and 5. the multivariate latent logistic model provides good approximation of the Bayesian network model which is well known good prediction model. This study implemented the proposed functions into the electro bullitein board and show the effectiviness. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Text Mining / Keywords Analysis / the multivariate latent logistic model / visualization system |
Paper # | ET2007-3 |
Date of Issue |
Conference Information | |
Committee | ET |
---|---|
Conference Date | 2007/5/12(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 | Educational Technology (ET) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Keywords analysis of electronic bulletin board using the multivariate latent logistic model |
Sub Title (in English) | |
Keyword(1) | Text Mining |
Keyword(2) | Keywords Analysis |
Keyword(3) | the multivariate latent logistic model |
Keyword(4) | visualization system |
1st Author's Name | Maomi UENO |
1st Author's Affiliation | The University of Electro-Communications() |
2nd Author's Name | Toshio OKAMOTO |
2nd Author's Affiliation | The University of Electro-Communications |
Date | 2007-05-19 |
Paper # | ET2007-3 |
Volume (vol) | vol.107 |
Number (no) | 48 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |