Basic Data Mining Concepts

CASE Table
A Case table has Attributes (columns) and States (values).
==> Equivalent to the Dimension Table in Data Warehousing
Case key = the PK of the table

NESTED Table
Nested case – the many part of the relationship.
==> Equivalent to the Fact Table in Data Warehousing
Nested key = Not the FK (i.e. the PK of the table). Important: It is rather the other attributes that make up the nested key.

Reference
Data Mining with SQL Server 2005 by ZhaoHui Tang and Jamie MacLennan

Data Mining 001

Definition:
Data mining is about analyzing data and finding hidden patterns using automatic or semi-automatic means.

Three Parts to Data Mining:
1. Create the model – similar to “create table”
discrete = distinct categories
continuous = numeric columns

2. Train the model – similar to “insert into table”
a. processing the model ==> similar to processing a cube
b. training model ==> truth table

3. Predict the model – similar to “select from table”
a. closing the analysis loop

Reference
Data Mining with SQL Server 2005 by ZhaoHui Tang and Jamie MacLennan