10. srpna 2012 15:41
I would like to do Shopping Basket analysis using only Excel and "Microsoft SQL Server 2008 Data Mining Add-ins for Microsoft Office 2007".
Is it possible to do that without connecting to "Analysis Services"? I dont have any access to database, i could provide the data in excel sheet and then Data Mining Add-ins would analyse it. Is it possible sth like that?
10. srpna 2012 18:17
No. It requires connection to SSAS.
You do not have to put data in SSAS. your data can be in Excel - But it needs connection to SSAS so that it can perform data mining wizardry.
Hope that helped.
10. srpna 2012 18:24Přispěvatel
1. Microsoft Data Mining Add-ins need to connect to Analysis Services. You can install trial version of SQL Server.
2. If you don't want to install SQL Server, you can use Predixion Insight (available from predixionsoftware.com). This product is similar to Microsoft Data Mining add-ins but connects to SQL Server in the cloud which is hosted by Predixion (you will need to pay for the service after 30-days)
Tatyana Yakushev [PredixionSoftware.com]
- Označen jako odpověď elchinibrahimov 10. srpna 2012 20:24
10. srpna 2012 20:29
Than you very much for your reply Paras and Tatyana,
But what is the purpose of connecting to Analysis Service? If i have data in Excel and there is add-in installed why this cant perform the analysis?
Do you know any other single software that would let me do shopping basket analysis without connecting to services and other things. I just want to use Excel as my raw data?
10. srpna 2012 20:47
Excel natively does not allow us to run data mining algo's (like running shopping basket analysis) - that's why we connect to data mining "engines" like SSAS or third party services like predixion (that Tatyana mentioned)
- Upravený Paras Doshi 10. srpna 2012 20:50
10. srpna 2012 21:06
So basically SSAS takes raw data from Excel and analysis it for patterns and returns the results to Excel again, right?
Do i have to have raw data in sql server?
10. srpna 2012 21:15
10. srpna 2012 21:24great thanks a lot Paras!