Potential of twitter archives

TU Graz - IICM

1

Thomas Altmann Graz, 3.7.2014 Potential of Twitter Archives

Potential of Twitter Archives

Thomas Altmann

TU Graz - IICM

2


“Twitter helps you create and share ideas and information instantly, without barriers.” (Twitter 2014)

Source: https://twitter.com/BarackObama/status/266031293945503744

TU Graz - IICM

3


Table of Contents

• Overview

• TweetCollector

• TwitterStat

• Demo

• Conclusion

TU Graz - IICM

4


Twitter

• Tweets

• Mentions

• Retweets

• Hashtags

TU Graz - IICM

5


Twitter Research

• Existing scientific research

– JAVA 2007, BOYD 2010

• Social network analysis

• Wealth of data

• Limited access

TU Graz - IICM

6


Research Questions

• What value can tweet archives provide?

• What value can the context of a tweet analysis provide?

TU Graz - IICM

7


Problems

• Retrieving old tweets

– Search: no older than six to nine days

– Users: no older than 3200 tweets

• Search does not find all tweets

– Only most popular

TU Graz - IICM

8


Twitter APIs

• REST API– Individual requests– Rate limited– Used for search and user timelines

• Streaming API– Single persistent connection– Set parameters– Constant real-time stream

TU Graz - IICM

9


TweetCollector - Purpose

• Collect tweets– Containing certain words/hashtags– From certain users

• Provide access to stored tweets– Basic web interface– Archives accessible through API

TU Graz - IICM

10


TweetCollector - Implementation

• Server-side– PHP/MySQL– Uses Twitter Streaming API and REST API– Provides API to access stored tweets

• Client-side– HTML/CSS/JavaScript– Create and delete archives– Start and stop archiving

TU Graz - IICM

11


Structure of Tools

Source: Own depiction

TU Graz - IICM

12


TwitterStat - Purpose

• Twitter Analysis Tool

• Analyze TweetCollector archives

• Show most active users, most used words/hashtags, …

• Provide context by linking original tweets

TU Graz - IICM

13


TwitterStat - Questions


TU Graz - IICM

14


TwitterStat - Answers

• Possible Answer

• Ordered List of results• Every result is a link

TU Graz - IICM

15


TwitterStat - Context

• Back to tweets

• Check original tweets responsible for results• Provide context for analysis

TU Graz - IICM

16


TwitterStat - API

• Access to analysis results

• Machine-readable format

• Possibility of further usage, e.g. visualization

TU Graz - IICM

17


TwitterStat - Visualization


TU Graz - IICM

18


LIVE DEMO

TU Graz - IICM

19


Results

• What value can tweet archives provide?– Access to tweets too old to be found by Twitter– Analysis needs large corpus of tweets– API to build applications using tweet archives

• What value can the context of a tweet analysis provide?– Determine content and sentiment of original tweets– Check if insights gained from analysis correspond with original

tweets– See tweets as part of a larger conversation

TU Graz - IICM

20


Conclusion

• Goal: Show potential of Twitter archives

• Developed: TweetCollector for archiving

• Developed: TwitterStat for analysis

• Showed: Value in tweet archives and analysis

TU Graz - IICM

21


Thank You!

TU Graz - IICM

22


References

A. JAVA, X. SONG, T. FININ, B. TSENG. Why we Twitter: Understanding Microblogging Usage and Communities. Presented at the Proceedings of the 9th WebKDD and 1st SNA-KDD 2007 workshop on Web mining and social network analysis (2007), 56–65.

D. BOYD, S. GOLDER, G. LOTAN. Tweet, Tweet, Retweet: Conversational Aspects of Retweeting on Twitter. Presented at the Proceedings of the 43rd Hawaii International Conference on System Sciences (2010), Hawaii.

Potential of twitter archives

Education

Transcript of Potential of twitter archives