Predictive Analytics

 

Goals: 

Introduction to the world of predictive analytics.

 

Techniques presented: 

Abandonment prevention, propensity to purchase, behavioral segmentation, basket analysis and credit risk, predictive and classification analyses.

 

Prerequisites: 

Minimum familiarity with the PC, with Windows and with the most common office automation tools is required.

 

Predictive analytics

• Introduction to predictive analytics

• CRISP-DM: a standard methodology for predictive analytics

 

Fields of analysis/case studies

• The prevention of abandonment

• The propensity to buy

• behavioral segmentation

• Basketball Analysis

• Credit risk and scoring grids

 

The mining process

• Discovering relationships between data

• Data sampling, partitioning and balancing

• The overtraining

• Evaluation of models

 

Predictive analytics and classification

• Decision trees

• The neural networks

• The linear regression

• Logistic regression

• K-means Clustering

• Two-step clustering

• Association Rules Apriori

 

Data mining workshops

• Examples of modeling through the analysis methodology