What kind of data do you content analyze?
My research group is working on a tool to apply natural language processing technology to content analyze large volumes of text. Our initial use case is looking for evidence of various group processes reflected in email messages of online groups (e.g., showing appreciation). However, we want to the tool to be generally useful and so are planning to extend the tool to handle other kinds of texts. We spent some time discussing what a reasonable next target would be (e.g., generic text files, web-based discussion groups, tweets), but I thought I should solicit opinions from other researchers. Hence the question: what kind of data are you content analyzing? Kevin Crowston Syracuse University Phone: +1 (315) 443-1676 School of Information Studies Fax: +1 (815) 550-2155 348 Hinds Hall Web: http://crowston.syr.edu/ Syracuse, NY 13244-4100 USA
Kevin, I am planning a content analysis of emails from climate change and environmental organizations, part of which will involve coding for the presence of certain climate-related frames. Just last night I handed in a proposal to that effect as part of an digital research methods class. One of the methods I considered was some sort of automated text analysis but given the relatively small sample--no more than 500 emails--I decided to go with a traditional content analysis instead. Hope that helps, Luis - - - - - Luis E. Hestres Ph.D. student | School of Communication | American University More about me at luishestres.com On Aug 9, 2012, at 9:22 PM, Kevin G Crowston <crowston@syr.edu> wrote:
My research group is working on a tool to apply natural language processing technology to content analyze large volumes of text. Our initial use case is looking for evidence of various group processes reflected in email messages of online groups (e.g., showing appreciation). However, we want to the tool to be generally useful and so are planning to extend the tool to handle other kinds of texts. We spent some time discussing what a reasonable next target would be (e.g., generic text files, web-based discussion groups, tweets), but I thought I should solicit opinions from other researchers. Hence the question: what kind of data are you content analyzing?
Kevin Crowston Syracuse University Phone: +1 (315) 443-1676 School of Information Studies Fax: +1 (815) 550-2155 348 Hinds Hall Web: http://crowston.syr.edu/ Syracuse, NY 13244-4100 USA
_______________________________________________ The Air-L@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
Join the Association of Internet Researchers: http://www.aoir.org/
Hi Kevin, I'd be interested using such as tool to analyse the "maintenance work" that is done in a social media community. I conducted a study, using manual content analysis, on comments in a large scale online community. I was attempting to categorise the sort of work that is done in the community, to maintain social order and ensure the community remains both an interesting and friendly place for new and existing users. We manually coded a sample set of comments, and developed a taxonomy. I was hoping to advance this work using text analytics to address the entire dataset, as, I imagine, the coded comments could be used as a training set. Let me know if this is along the lines of what you are thinking. Rgds, j On 10 August 2012 02:22, Kevin G Crowston <crowston@syr.edu> wrote:
My research group is working on a tool to apply natural language processing technology to content analyze large volumes of text. Our initial use case is looking for evidence of various group processes reflected in email messages of online groups (e.g., showing appreciation). However, we want to the tool to be generally useful and so are planning to extend the tool to handle other kinds of texts. We spent some time discussing what a reasonable next target would be (e.g., generic text files, web-based discussion groups, tweets), but I thought I should solicit opinions from other researchers. Hence the question: what kind of data are you content analyzing?
Kevin Crowston Syracuse University Phone: +1 (315) 443-1676 School of Information Studies Fax: +1 (815) 550-2155 348 Hinds Hall Web: http://crowston.syr.edu/ Syracuse, NY 13244-4100 USA
_______________________________________________ The Air-L@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
Join the Association of Internet Researchers: http://www.aoir.org/
-- John McAuley ------ Knowledge & Data Engineering Group Trinity College Dublin Ireland ------
participants (3)
-
John McAuley -
Kevin G Crowston -
Luis Hestres