-
Book Overview & Buying
-
Table Of Contents
Data Analysis Using SQL and Excel - Second Edition
By :
A data mining model takes inputs and produces an output, which is typically a prediction or estimation of some value. There are two major processes involved with models. The first is training or building the model. The second is applying the model to new data.
SQL provides a good basis for learning the basics about data mining. Although this may seem surprising, some powerful techniques are really more about manipulating data than about fancy statistical techniques. The GROUP BY operation in SQL is analogous to creating a model (both summarize data). The JOIN operation is analogous to scoring a model.
This chapter discusses several different types of models. The first is a look-alike model, where the model score indicates how close one example is to another. For instance, the model score might indicate how similar zip codes are to the zip code that has highest market penetration.
Lookup models are another type. These create a lookup table, so the process of scoring the...
Change the font size
Change margin width
Change background colour