Unsupervised feature selection for linked social media data
This presentation is the property of its rightful owner.
Sponsored Links
1 / 9

Unsupervised Feature Selection for Linked Social Media Data PowerPoint PPT Presentation


  • 176 Views
  • Uploaded on
  • Presentation posted in: General

Unsupervised Feature Selection for Linked Social Media Data. Jiliang Tang, Huan Liu Arizona State University. Motivations. Feature selection is effective in dealing with high-dimensional data.

Download Presentation

Unsupervised Feature Selection for Linked Social Media Data

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.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Unsupervised Feature Selection for Linked Social Media Data

Jiliang Tang, Huan Liu

Arizona State University


Motivations

  • Feature selection is effective in dealing with high-dimensional data.

  • Absence of label information associated with the features. (Traditional unsupervised feature selection algorithms fail.)

  • Link information in the social media data could be useful.


Linked Data in Social Media


Linked Unsupervised Feature Selection(LUFS)


Experiments


Source Free (proposal)

- Open to discussion


General Framework


2-Step iterative learning

  • Learn for the labeled training data & unlabeled data.

  • Retrieve the unlabeled data from open database, such as Web etc.


Preliminary studyon text classification tasks


  • Login