Sun 30 Oct 2005
Business Intelligence/Data-mining/Analytics
Posted by Nishith under Open Source Analytics[2] Comments
If you are doing business intelligence/analytics/data-mining today and are satisfied with your results, this blog may not be for you.
This blog is about doing the best of that stuff in an accessible (here) and affordable (free) way. Its about Open Source Analytics.
February 15th, 2007 at 2:32 am
Business intelligence is the technology and practice of applying information to make decisions. In this way, business intelligence is different than data warehousing, which is more about storing information. It is also more comprehensive than data mining. Information shows its real value when many people can use and share it. This is the goal of business intelligence.
For information to be this usable, it must be trusted, timely, relevant, easy-to-use, and in context. These are all necessary aspects of business intelligence technology. Different business intelligence tools address these factors in different ways. Reporting delivers regular, timely information, with the ability to author reports or queries to get specific details. OLAP analysis, with its multiple dimensions, allows you to compare and contrast information against time and other factors to uncover trends. Scorecarding presents your key performance metrics and whether you’ve cross pre-determined thresholds. Executive dashboards put information in context, and in an easy-to-understand format.
February 16th, 2007 at 9:41 am
Thank Chris for your comments. You are right, there are so many buzz words and a complicated terminology out there that ends up confusing most people. The marketing hype only confounds things further.
You are right that Business Intelligence is by definition more comprehensive than data mining as it also covers timely information delivery in simplified and usable forms to end users.
However, the term BI seems to have been misappropriated by the commercial vendors to mean the set of specific features their products cover. For example, statistical modeling (part of data-mining) seems to be completely left out from most BI definitions out there. It’s quite weird, really!