03-19-2014 04:06 PM
I would like to invite you to my e-poster presentation where I am going to show companies can benefit from including in-house and Internet scattered text sources to Business Intelligence. I will explain how to extract data from different source, next clean and process them, perform Text Mining analysis and Sentiment analysis of text in other language than English using SAS Text Miner 12.1 and finally apply obtained variables from sentiment analysis to sales analysis.
The biggest challenge is to match comments found in the Internet with sales of a given product. I will also provide you with two solutions to this problem.
In case you cannot attend my presentation and encourage you to read my paper after SGF 2014 titled Text Mining Reveals the Secret of Success: Identification of Sales Determinants Hidden in Customers' Opinions. You will find there description of methods I used along with the macros and build.
I am looking forward to your feedback and comments.
03-31-2014 07:49 PM
in case you could not attend my presentation I provide you with link to my paper: http://support.sas.com/resources/papers/proceedings14/1652-2014.pdf.
I hope everyone will find something interesting for themselves. For some it can be macros, for some analytics approach and for some just theory
If you have any questions or want to provide me with feedback there is my email in the Contact section.
I am looking forward to your questions and feedback.
04-01-2014 11:31 AM
Hi Rafał, This looks like a great resource. Thanks for sharing! Best, Anna-Marie
04-02-2014 05:05 PM
it is nice to hear you like my paper. If you have any feedback I will be glad to read
04-01-2014 01:21 PM
Interesting. Anyone know if the e-posters themselves are available somewhere (or will be?).
04-02-2014 05:11 PM
it is nice to hear you like my paper. Regarding eposter last year's proceedings were organised this way https://support.sas.com/resources/papers/proceedings13/#Posters
However, as I see this years you cannot filter by presentation type but you can find eposters there if you know the titles. For example, this one done by my collegues is really interesting: http://support.sas.com/resources/papers/proceedings14/1883-2014.pdf