Home > Data Mining and Oracle Data... > How Do You Mine Data? > Understanding Data and Mini... > Preparing the Data > Data Preparation Specifical...
You may have to modify the data before you can use it to build models.
Even if the data is in an Oracle data warehouse, you may be necessary to join tables to create an input table for a mining build. For example, sales data and demographic data may be in different tables.
The attribute that you plan to predict may not exist in the data set. For example, you may want to predict the value of the binary attribute ATTRITE
, with values 0 and 1. The data set might contain an attribute ACCOUNTCLOSE
with the date when the account was closed.
You may want to define attributes for business reasons. For example, you may want to define an attribute that identifies the sales region.
You also must understand the data that you plan to use by calculating statistics, such as maximum, minimum, mean, and standard deviation. It is useful to look at the distribution of attribute values, for example, to view histograms. The Explore Data Node helps perform this task by calculating statistics and creating histograms for a table or view in the database.
Oracle Data Miner calculates statistics and generates graphical displays to visualize data.
Oracle Data Miner enables you to create transformations and manage them.