1. 1
W I S S E N  T E C H N I K  L E I D E N S C H A F T
 www.tugraz.at
Analysis and Visualization
of Real-Time Twitter Data
19.11.2015
Sead Harmandic

2. 22
Motivation
• Social Network and Networking
• Micro-Blogging
• Twitter
• Launched in 2006
• Active users per month
• ~ 316 Milions (August)
• ~ 320 Milions (current)
• Tweets per day ~ 500 Milions
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

3. 33
Problem and Research Objective
• Problems with Twitter
• Event based data
• Detail event information
• Collection of information
• Research Objective
• What kind or sort of information are we capable of
providing during and after Twitter event?
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

4. 44
State of the Art I
Analyse Twitter as a form of electronic word-of-mouth in
correlation to brands and the influence of the service on
various brands. [Jansen et al., 2009]
• Brands
• H&M, Honda, Exxon, Dell, Lenovo, Amazon, etc.
• Opinion (sentiment)
• None; Wretched ; Bad; So-So; Swell; Great
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

5. 55
State of the Art II
Using Twitter and (classified) real-time data in order to
notify the public about the eathquake. [Sakaki et al.,
2010]
• Test region: Japan
• Large ammount of Twitter users
• High rate of earthquakes per year
• Twitter user  sensor
• Tweet  sensor information (social sensor)
• Toretter („we have taken it“) since 2010
• Faster then Japan Meteorogical Agency
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

6. 66
Available Tools
• TweetTracker
• Pros: Geo. Maps; Translation of Non-English; Keyword
comparison
• Cons: Visualizing up to 7500 Tweets
• TweetArchivist
• Pros: Top Users; Top Hashtags; Language
• Cons: No storage or APIs, Paid service
• twExplorer
• Pros: Top Users; Top Hashtags
• Cons: No archiv or APIs, Maximum of 500 Tweets
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

7. 77
TwitterSuitcase
• Why Suitcase
• Identification
• Objective
• TU Graz Twitter Applications
• TweetCollector (raw Twitter data)
• TwitterWall (event representation)
• TwitterStat (analysis of keyword, hashtag or person)
• TweetGraph (scope of tweets)
• TwitterSuitcase
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

8. 88
TwitterSuitcase - Concept
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

9. 99
TwitterSuitcase – Overview
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

10. 1010
TwitterSuitcase – Categories I
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Top Users

11. 1111
TwitterSuitcase – Categories II
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Top Links

12. 1212
TwitterSuitcase – Categories III
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Most Popular Retweets

13. 1313
TwitterSuitcase – Categories IV
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Timeline

14. 1414
TwitterSuitcase – Categories V
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Top Words

15. 1515
TwitterSuitcase – Categories VI
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Top Software

16. 1616
TwitterSuitcase – Categories VII
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Most Popular Hashtags

17. 1717
TwitterSuitcase – Categories VIII
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Top Screenshots

18. 1818
TwitterSuitcase – Categories IX
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC
• Wikipedia Article(s)

19. 1919
TwitterSuitcase – Use Case I
• European Massive Open Online Courses
• #emoocs2014
• Total of 4450 Tweets
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

20. 2020
TwitterSuitcase – Use Case II
• Most Popular Hashtags
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

21. 2121
TwitterSuitcase – Use Case III
• Top Users
• moocf(185), Agora Sup(141), fuscia info(134), pabloachard(124),
mooc24(120), tkoscielniak(103), bobreuter(85), OpenEduEU(84),
yveszieba(81), redasadki(81) ~ 25%
• Top Link(s)
• http://bit.ly/1la3yJX (32)  HTML Page „eLearning Papers Issue 37“
• Top Words
• RT(2567), moocs(802), mooc(639), learning(339) and openedueu(319).
The rest of the words belong mostly to prepositions or articles.
• Used Software
• Web(1574 or 35.4%), Apple devices(1253 or 28.5%), TweetDeck(564 or
12.7%), Android devices(288 or 6.5%)
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

22. 2222
Conclusion
• TwitterSuitcase
• Research objective  What kind or sort of information are we capable of
providing during and after Twitter event?
• TwitterSuitcase extensions
• Visualizing Tweets on Geographical Maps; Region-Tweet-Search
• MentionMaps
• ReTweets; HTTP Links; Data sources; etc.
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

23. 2323
Thank you for your attention.
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC

24. 2424
Bibliography
[Java et al., 2007] Java, A., Song, X., Finin, T., and Tseng, B. (2007). Why we twitter:
Understanding microblogging usage and communities. Proceedings of the 9th WebKDD
and 1st SNA-KDD 2007 workshop on Web mining and social network analysis, pages 56–
65.
[Jansen et al., 2009] Jansen, B. J., Zhang, M., Sobel, K., and Chowdury, A. (2009). Twitter
power: Tweets as electronic word of mouth. Journal of the American society for information
science and technology, page 2169–2188.
[Sakaki et al., 2010] Sakaki, T., Okazaki, M., and Matsuo, Y. (2010). Earthquake shakes
twitter users: real-time event detection by social sensors. Proceedings of the 19th
international conference on World wide web, pages 851–860.
Analysis and Visualization of Real-Time Twitter Data
19.11.2015
Sead HARMANDIC