Thursday, January 03, 2013 2:15 PM
i want to do sentiment analysis on my own data, which results the summary about the phrase like (bad, neutral, good).
i have done this before in IBM text analytics for surveys software. But want to know that microsoft have some sort of algorithm like this.
we have retail client, who wants to know the sentiments of the consumers about there products. that either it is positive, negative or neutral.
please help me on this.
Thursday, January 03, 2013 5:28 PMAnswererSentiment Analysis is not part of SQL Server. I don't know if there are any other products from Microsoft that allow you to do this.
Tatyana Yakushev [PredixionSoftware.com]
- Proposed As Answer by Paras Doshi Thursday, January 03, 2013 5:53 PM
Thursday, January 03, 2013 5:52 PM
Sometime back, MS ran a beta project codenamed "Social Analytics" - it's not open anymore - But I hope it does sometime in future. So while currently, we do not have something out of the box for this, I hope we have it in future. A blog I did on the project: http://parasdoshi.com/2012/06/11/this-is-cool-microsoft-project-codename-social-analytics/
And by any chance are you considering building a custom model using SSAS Data Mining algorithms?
I am thinking we can use "MS Logistic regression" algo. But just to get comments from experts, I posted a thread here: http://social.msdn.microsoft.com/Forums/en-US/sqldatamining/thread/6c95416a-05bf-4a63-a64f-ed97645fdb3e
- Marked As Answer by RafayMirza Friday, January 04, 2013 12:28 PM
Friday, January 04, 2013 11:54 AM
Thanks paras doshi.
yes i am trying to build a model in SSAS which help us doing sentiment analysis on reviews of customers.
I have already look towards beta project codenamed "Social Analytics", but it's not useful anymore.