Little note on data mining

Little note on data mining

 

Data mining program

1. Weka is the free tool to do the data mining

 

Data mining methodology

CRISP-DM (cross-industry standard process for data mining)

1.    Project objective

2.    Data understanding; what is the meaning of each variable?,

a.    data exploration – learning the relationship of the data in each group

b.    data selection

3.    Data preparation

-    Making the data more unify or transforming the data to the format which could be used for modeling

-    Cleaning the data, I think

4.    Modeling

a.    Using algorithm to create the model

5.    Evaluation

a.    Which feature is important?

b.    Which feature affect the prediction?

6.    Deployment

a.    Testing the prediction model

 

Ref figure: https://commons.wikimedia.org/w/index.php?curid=24930610

 

Info source note:

YouTube channel: data professor