Comments on: Web Mining: Short/Long Term User Profile http://www.dataminingblog.com/web-mining-shortlong-term-user-profile/ Data mining crossroads - research, applications, news, list of blogs and customized search engine about data mining. Tue, 07 Feb 2012 08:14:55 +0100 http://wordpress.org/?v=2.8.5 hourly 1 By: Intelestream Inc http://www.dataminingblog.com/web-mining-shortlong-term-user-profile/comment-page-1/#comment-76906 Intelestream Inc Fri, 16 Apr 2010 05:09:27 +0000 http://www.dataminingblog.com/?p=926#comment-76906 Good article! I agree that CRM is becoming more and more accessible to small businesses as a Web-based application, thanks to the fact that in-house IT professionals are no longer required. Now, applications are accessible online - all you need is a Web browser and internet connection. Our online application, intelecrm (www.intelestream.net/intelecrm), focuses on small and medium sized businesses and provides a unique flat fee pricing model for unlimited users. Anyone shopping for a powerful and affordable CRM application is encouraged to try it for free. Good article! I agree that CRM is becoming more and more accessible to small businesses as a Web-based application, thanks to the fact that in-house IT professionals are no longer required. Now, applications are accessible online – all you need is a Web browser and internet connection. Our online application, intelecrm (www.intelestream.net/intelecrm), focuses on small and medium sized businesses and provides a unique flat fee pricing model for unlimited users. Anyone shopping for a powerful and affordable CRM application is encouraged to try it for free.

]]> By: Sandro Saitta http://www.dataminingblog.com/web-mining-shortlong-term-user-profile/comment-page-1/#comment-75349 Sandro Saitta Mon, 22 Feb 2010 21:44:43 +0000 http://www.dataminingblog.com/?p=926#comment-75349 @Christian: Thanks for your interest. In my models, there are always variables from both side (CRM and web) which are selected. I have tried model based only on CRM data (resp. web data) but accuracy is not as good as a combination of both kind of data. Regarding the number of variables selected in my decision tree, it is the same proportion as the input data (3/4 CRM data, 1/4 for web behavior). @Christian: Thanks for your interest. In my models, there are always variables from both side (CRM and web) which are selected. I have tried model based only on CRM data (resp. web data) but accuracy is not as good as a combination of both kind of data. Regarding the number of variables selected in my decision tree, it is the same proportion as the input data (3/4 CRM data, 1/4 for web behavior).

]]> By: Christian http://www.dataminingblog.com/web-mining-shortlong-term-user-profile/comment-page-1/#comment-75348 Christian Mon, 22 Feb 2010 21:15:54 +0000 http://www.dataminingblog.com/?p=926#comment-75348 Hi Sandro, which are the better key drivers for your click-targeting. The sociodemographic's from the CRM database or the affinities in web usage? Regards, Christian Hi Sandro,

which are the better key drivers for your click-targeting. The sociodemographic’s from the CRM database or the affinities in web usage?

Regards, Christian

]]>