1 / 13

Term Weighting Schemes for Question Categorization

Presenter : Cheng-Han Tsai Authors : Xiaojun Quan , Wenyin Liu, Senior Member, IEEE, and Bite Qiu TPAMI, 2012. Term Weighting Schemes for Question Categorization. Outlines. Motivation Objectives Methodology Experiments Conclusions Comments. Motivation. Text categorization.

mary-beard
Télécharger la présentation

Term Weighting Schemes for Question Categorization

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Presenter : Cheng-Han Tsai Authors : XiaojunQuan, Wenyin Liu, Senior Member, IEEE, and Bite Qiu TPAMI, 2012 Term Weighting Schemes for Question Categorization

  2. Outlines Motivation Objectives Methodology Experiments Conclusions Comments

  3. Motivation Text categorization Questions are usually a piece of short text, can the existing term-weighting methods perform consistently in question categorization as they do in text categorization?

  4. Objectives UIQA This paper proposed new supervised term-weighting methods to deal with the problems that questions are usually a piece of short text

  5. Methodology : :

  6. Methodology

  7. Methodology

  8. Experiments – different values of k 20 Newsgroups Yahoo-natural Yahoo-500 Yahoo-1000

  9. Experiments – different kernels Yahoo-500 Yahoo-1000

  10. Experiments – different scales of data Yahoo-500 Yahoo-1000

  11. Experiments – overall Statistical Significance Test10-fold cross-validation Value of k: 30 Kernel: LINEAR

  12. Conclusions The three new methods, especially iqf*qf*icf, exhibit stable and consistent improvement over most of the previous term-weighting methods mentioned in the experiments.

  13. Comments • Advantages • This paper compares to many well-known methods, and it performs well. • Applications • Question Categorization

More Related