Text Mining and Social Network Visualizations

Text Mining and Network Visualizations of Online Communities

In this project we are developing and evaluating various tools and techniques to automatically uncover the social networks of online participants from their digital footprints alone. One of the tools that we are developing is Netlytic, a cloud-based text and social networks analyzer that can automatically summarize large volumes of text and discover social networks from conversations on social media sites such as Twitter, YouTube, blogs, online forums and chats. With Netlytic, you can:

  1. capture (or import) online conversational type data such as tweets, blog comments, forum postings and text messages, etc…(eg…build and collect your own unique data set or import in your own existing data set)
  2. find and explore emerging themes of discussions among individuals within your data set,
  3. build and visualize communication networks to discover and explore emerging social connections between individuals within online communities.

Netlytic can automatically build and visualize two types of communications network: 1) a chain network based on who-replies-to-whom and 2) a personal name network based on who-mentioned-whom.  It is ideally suited for analyzing online interactions within any large online communities such as Twitter, fan/discussion forums, customer review forums, online classes, health support groups, etc… Specifically, Netlytic can be used to automatically discover what people within an online community are talking about, who is talking to whom, how often they are communicating, the nature of their relationships or interactions (are community members happy, friendly and supportive; or are they angry, hostile and disrespectful to each other) and relatively how strong their relationships are. Once discovered, social network information can be used in a myriad of ways such as measuring the strength of online communities, identifying and analyzing consumers’ perceptions of products and services, finding popular resources and sharing information within a network of trust, and assessing the effectiveness of social media marketing campaigns.

Netlytic has successfully been used  to analyze a wide variety of online communities and networks including: learning communities [1,2], communities of bloggers and blog readers [3,4], communities emerging on the i-Neighbors website [5], a scholarly community on Twitter [6] and most recently a community of Tolkien’s fan on the the popular website, TheOneRing.net, a website dedicated to discussions of Tolkien’s (text) and Peter Jackson’s (film) versions of The Lord of the Rings [7-8].

Footnotes:

[1] Caroline Haythornthwaite and Anatoliy Gruzd. “Analyzing Networked Learning Texts.” Proceedings of Networked Learning Conference (2008): 136-143;

[2] Anatoliy Gruzd, “Studying Collaborative Learning Using Name Networks.” Journal of Education for Library and Information Science 50, no. 4 (2009): 243-253;

[3] Anatoliy Gruzd, “Automated Discovery of Emerging Online Communities Among Blog Readers: A Case Study of a Canadian Real Estate Blog,” (conference paper, Internet Research 10.0 Conference, Milwaukee, WI, October 7-11, 2009);

[4] Chung Joo Chung, Anatoliy Gruzd, and Han Woo Park, “Developing an e-Research Tool for Humanities and Social Sciences: Korean Interent Network Miner on Blogosphere,” Journal of Humanities 60, no.12 (2010): 429-446.

[5] Keith N. Hampton, “Internet Use and the Concentration of Disadvantage: Glocalization and the Urban Underclass,” American Behavioral Scientist 53, no. 8 (2010): 1111-1132. http://abs.sagepub.com/content/53/8/1111

[6] Anatoliy Gruzd, Yuri Takhteyev, and Barry Wellman. “Imagining Twitter as an Imagined Community.” American Behavioral Scientist, Special Issue on Imagined Communities 55, no. 10 (2011): 1294-1318. http://abs.sagepub.com/content/55/10/1294

[7] Jennifer Grek Martin, “Two Roads to Middle-earth Converge: Observing Text-based and Film-based Mental Images from TheOneRing.net Online Fan Community.” (master’s thesis, Dalhousie University, 2011). http://dalspace.library.dal.ca/handle/10222/14242.

[8] Martin, J.M.G., Gruzd, A., Howard, V. (2013). Navigating an imagined Middle–earth: Finding and analyzing text–based and film–based mental images of Middle–earth through TheOneRing.net online fan community.  First Monday18(5 – 6). DOI: 10.5210%2Ffm.v18i5.4529