Studying online conversations in the Korean blogosphere: A network approach <ul><li>Anatoliy Gruzd  (gruzd@dal.ca) </li></...
What content-based features of online interactions help to uncover nodes and ties between online participants? Automated D...
ICTA - Online Tool for Social Network Discovery  http://TextAnalytics.net
Objectives <ul><li>Automated analysis of online interactions and conversations in the Korean blogosphere </li></ul><ul><li...
Dataset <ul><li>OhMyNews – popular blogging website in Korea </li></ul><ul><li>Single blog authored by  방짜  (bangzza)  htt...
Sample blog post and comments
Automated Discovery of Social Networks Chain (reply-to) method Visualized by CMU ORA
Automated Discovery of Social Networks    Name Network   Approach FROM: Ann “ Steve  and  Natasha , I couldn't wait to see...
<ul><li>Main Communicative Functions of Personal Names (Leech, 1999)   </li></ul><ul><ul><li>getting attention and identif...
Network representation of blog comments
Semi-automated network evaluation 1   난  (I)  2   사진쟁이  (Photographer)  3   그래서  (and, so) 4   테츠  (Tetz) 5   방짜  (Bangzza...
Clues suggesting that a word is likely to be a nickname <ul><li>context words such as &quot; 님 &quot; = an honorific or &q...
Words that are NOT likely to be used as a nickname <ul><li>a word candidate is a phrase  </li></ul><ul><ul><li>e.g., if th...
Words that are NOT likely to be used as a nickname (2) <ul><li>a word candidate appears more than one time in the comment ...
Conclusion <ul><li>A network representation of comments posted to a blog makes it much easier to analyze social interactio...
Acknowledgments  <ul><li>Jaeeun Yoo at the University of Toronto for her help with the data analysis  </li></ul><ul><li>Th...
Upcoming SlideShare
Loading in...5
×

Studying online conversations in the Korean blogosphere: A network approach

701

Published on

Studying online conversations in the Korean blogosphere: A network approach

Published in: Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
701
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
10
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • .
  • Studying online conversations in the Korean blogosphere: A network approach

    1. 1. Studying online conversations in the Korean blogosphere: A network approach <ul><li>Anatoliy Gruzd (gruzd@dal.ca) </li></ul><ul><li>Dalhousie University, Canada </li></ul><ul><li>Chung Joo Chung (cjchung@buffalo.edu) </li></ul><ul><li>State University of New York at Buffalo, USA </li></ul><ul><li>Han Woo PARK (hanpark@ynu.ac.kr) </li></ul><ul><li>YeungNam University, Korea </li></ul>International Sunbelt Social Network Conference Riva del Garda, Italy July 3, 2010
    2. 2. What content-based features of online interactions help to uncover nodes and ties between online participants? Automated Discovery of Social Networks among Blog Readers/Commentators ?
    3. 3. ICTA - Online Tool for Social Network Discovery http://TextAnalytics.net
    4. 4. Objectives <ul><li>Automated analysis of online interactions and conversations in the Korean blogosphere </li></ul><ul><li>Develop and evaluate an e-research tool called the Korean Internet Network Miner (KINM) </li></ul>
    5. 5. Dataset <ul><li>OhMyNews – popular blogging website in Korea </li></ul><ul><li>Single blog authored by 방짜 (bangzza) http://blog.ohmynews.com/bangzza </li></ul><ul><li>~1,000 comments (April 2008 - April 2009) </li></ul>
    6. 6. Sample blog post and comments
    7. 7. Automated Discovery of Social Networks Chain (reply-to) method Visualized by CMU ORA
    8. 8. Automated Discovery of Social Networks Name Network Approach FROM: Ann “ Steve and Natasha , I couldn't wait to see your site. I knew it was going to [be] awesome!” This approach looks for personal names in the content of the comments to identify social connections between online participants. Method Connect the sender to people mentioned in the message Connect people whose names co-occur in the same message(s) Discovered Tie(s) Ann -> Steve Ann -> Natasha Steve <-> Natasha
    9. 9. <ul><li>Main Communicative Functions of Personal Names (Leech, 1999) </li></ul><ul><ul><li>getting attention and identifying addressee </li></ul></ul><ul><ul><li>maintaining and reinforcing social relationship s </li></ul></ul><ul><li>Names are “one of the few textual carriers of identity” in discussions on the web (Doherty, 2004) </li></ul><ul><li>Their use is crucial for the creation and maintenance of a sense of community (Ubon, 2005) </li></ul>Automated Discovery of Social Networks Name Network Approach
    10. 10. Network representation of blog comments
    11. 11. Semi-automated network evaluation 1 난 (I) 2 사진쟁이 (Photographer) 3 그래서 (and, so) 4 테츠 (Tetz) 5 방짜 (Bangzza) 6 댓글 (comment) 7 녹두 (Nokdu) 8 ㅋㅋ (  , : )) 9 좀 (a little, a bit) 10 사람 (people) Among 10 nodes, only 2 , 4 , 5 and 7 are NANEs or IDs of participants in the Bangzza blog 1 2 3 5 6 7 10 8 9 4
    12. 12. Clues suggesting that a word is likely to be a nickname <ul><li>context words such as &quot; 님 &quot; = an honorific or &quot; 씨 &quot; = Mr./Ms </li></ul><ul><li>full name, which is almost always three characters </li></ul><ul><li>punctuation indicative of someone being addressed (e.g., “/” or “:”) </li></ul><ul><li>combination of characters (Korean, English and/or Chinese), symbols (e.g., underscores, hyphens) and numbers </li></ul><ul><li>patterns indicative of non-native words </li></ul><ul><ul><li>phonetic koreanization of English (e.g., &quot; 미디어몽골 &quot; = mediamogul = Media Mogul) </li></ul></ul><ul><ul><li>phonetic romanization of Korean (e.g. “jihwaja” = 지화자 ) </li></ul></ul>
    13. 13. Words that are NOT likely to be used as a nickname <ul><li>a word candidate is a phrase </li></ul><ul><ul><li>e.g., if the “FROM” field is used more like a subject line (possible indicators include white spaces and length) </li></ul></ul><ul><li>a word candidate consists of a single character (e.g., “a” or “ ㄱ ” ) </li></ul><ul><li>a word candidate consists of netspeak </li></ul><ul><ul><li>emoticons (e.g. “=_=”) </li></ul></ul><ul><ul><li>slang and abbreviations (e.g., using “2MB” to refer to the former Korean president) </li></ul></ul><ul><ul><li>onomatopoeia (e.g., &quot; ㅋㅋ ” = heehee, &quot; 하하 ” = haha) </li></ul></ul>
    14. 14. Words that are NOT likely to be used as a nickname (2) <ul><li>a word candidate appears more than one time in the comment </li></ul><ul><li>a word candidate consists of random characters (e.g. &quot; ㅁㄴㅇㄹ &quot; or “asdf”) </li></ul><ul><li>a word candidate is a short, conversational word or phrase (e.g., &quot; 나 &quot; = me,&quot; 아이고 &quot; = oh no, &quot; 그래서 &quot; = so/therefore) </li></ul><ul><li>a word candidate is a common word or idea in the given context/topic (e.g &quot; 대한민국 &quot; = Republic of Korea, &quot; 쥐체사상 &quot; = a newly created word used to refer to political fanatics) </li></ul>
    15. 15. Conclusion <ul><li>A network representation of comments posted to a blog makes it much easier to analyze social interactions among online participants </li></ul><ul><li>Even in a blog dominated by mostly anonymous and argumentative commentators, a community can still emerge </li></ul><ul><li>Suggested future improvements to our network discovery algorithm. </li></ul>
    16. 16. Acknowledgments <ul><li>Jaeeun Yoo at the University of Toronto for her help with the data analysis </li></ul><ul><li>The project is in part supported by </li></ul>
    1. A particular slide catching your eye?

      Clipping is a handy way to collect important slides you want to go back to later.

    ×