Difference between revisions of "Analyzing URL Chatter on Twitter"

From Knoesis wiki
Jump to: navigation, search
Line 32: Line 32:
 
*2. Themes-Entity extraction from the tweets rather than using the present available themes
 
*2. Themes-Entity extraction from the tweets rather than using the present available themes
 
*3. Provisions in the DB to know the popularity of the url at that particular theme.
 
*3. Provisions in the DB to know the popularity of the url at that particular theme.
 +
===Assumptions===
 +
* URL max length in the DB is 300

Revision as of 19:28, 9 November 2009

Project Description

This project helps analyzing the urls available in the tweets with the theme. The data(Tweets) crawled for twitris project is being used.

Objectives

  • Classify the content of the urls
  • Understand user perception of the websites

Motivation

  • Search Engine Perspective - How to choose a page which is interesting to the user, given the keywords
  • Publisher Perspective - What do people think about the page(URL)

Status

Week 1

  • Extracting the Urls from the tweets - Done
  • Recognizing the short/tiny Urls and transforming it into the long Urls. - Done

Week 2

  • Creating a table for the Urls and the tweets to store the urls - In Progress
  • Analysing the urls with the presently available themes - In Progress

Week 3

Week 4

Future work

  • 1. Make sure the Url is not short by checking it recursively
  • 2. Themes-Entity extraction from the tweets rather than using the present available themes
  • 3. Provisions in the DB to know the popularity of the url at that particular theme.

Assumptions

  • URL max length in the DB is 300