Data mining specific steps

Data mining specific steps

1. Understand business and data

2. Prepare data

Data cleaning:

Missing value handling:

Outliers:

Data normalization:

Feature selection:

Data sampling processing:

3. Data modeling

Classification question:

Clustering problem:

regression problem

Correlation Analysis

Ensemble learning

image
Bagging (such as random forest algorithm)
Boosting
Stacking

4. Model evaluation

Confusion matrix and accuracy index

Generalization ability assessment

Other models:

Evaluate data processing:

5. Application

Model saving:
Model optimization: