Can you please give me a simple definition of 1. Confidence 2. Support used in Data Mining/Intelligence These are...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
fairly ubiquitous words in and out of the spaces of DM/BI/mining, but confidence can refer to the anticipated range of an output variable given a set of input variable values. The confidence interval (CI) establishes the range (high sales level +/- CI) within which the corresponding predicted output (i.e., sales) per combination of the inputs (i.e., demographics) occurs 95% of the time. The higher the CI, the less predictive, and therefore useful, the model. When models are tested against broad real-world company data, the model itself is tested. CIs are used as a key indicator of a model's performance. Confidence intervals can be used to determine which inputs, i.e., combinations of demographic attributes, have the most predictive effect on the outputs, i.e., sales. At least this is how I've used the term on occasion. This may or may not apply to your situation. If you can provide more context on your situation, that would be helpful.
For more information, check out searchCRM's Best Web Links on Data Mining.
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.