To determine the Influential Users of a social network and their interests, many researchers recommend exploring and mining user's activity especially what they write. Although they have introduced efficient and accurate methods for text mining in social networks, here are some ideas by which the efficiency and accuracy can be enhanced. To evaluate our idea we will need to some benchmark data sets. Could you please recommend some free data sets for evaluating text mining in social networks?
A promising place seems to be http://trec.nist.gov/data/tweets/, where a significant sample of tweets collected between Jan 23th and Feb 8th 2011 is available for download.
The Library of Congress received from Twitter a copy of the corpus of tweets as of 2010. You can find some reference here: http://blogs.loc.gov/loc/2010/04/the-library-and-twitter-an-faq/, but you should verify if data are accessible,
We have collected a corpus of tweets in our free Web-based twitter mining tool available from http://tweetminer.eu . Via the interface you can access the corpus via Booleans operators and your definitiion of location and time. Detailed descriptions of the data set and the free Web serivce are available from our paper at http://personalwebpages.deusto.es/reips/pubs/papers/2011ReipsGaraizar_final.pdf The reference is
Reips, U.-D., & Garaizar, P. (2011). Mining Twitter: Microblogging as a source for psychological wisdom of the crowds. Behavior Research Methods, 43, 635-642.
A good repository is the Stanford Large Network Dataset Collection
at: http://snap.stanford.edu/data/
- Badges
- Science topic