Hi Ramprakash, You may find this paper useful. http://www.linkedin.com/news?viewArticle=&articleID=831117876&gid=115439&type=member&item=74678437&articleURL=http%3A%2F%2Fwww.slideshare.net%2Fsatyajit_gupte%2Fsentiment-mining-engine-architecture&urlhash=7IWn&trk=group_items_see_more-0-b-shrttl
On Thu, Dec 22, 2011 at 11:12 AM, Ramprakash Ramamoorthy < [email protected]> wrote: > On Thu, Dec 22, 2011 at 12:24 AM, Isabel Drost <[email protected]> wrote: > > > On 16.12.2011 Ramprakash Ramamoorthy wrote: > > > Thank you all for your replies. > > > > > > As Sreejith suggested, I am going ahead with creating a data set in my > > > domain itself(Customer Support). I don't find any classified data sets > in > > > this aspect. > > > > Not quite your domain but maybe some ideas from the following publication > > might > > help you generate an initial candidate set for training: Enhanced > Sentiment > > Learning Using Twitter Hashtags and Smileys > > > > > > Isabel > > > > @Isabel, > > Tried my luck with it already. But it was all about colloquial > slang and short interactions. Where as in the tech support domain(say a > CRM), communication is very much official and formal. There is a huge > difference between the way official mails are written and tweeting is done. > > -- > With Thanks and Regards, > Ramprakash Ramamoorthy, > B.Tech ICT, > SASTRA University. > +91 9626975420 > -- *Sreejith.S* http://srijiths.wordpress.com/ * *http://sreejiths.emurse.com/ tweet2sree@twitter <http://tweet2Sree>
