rodolfolermacontreras / supreme-octo-spork

Ensemble Models in ML

supreme-octo-spork

Ensemble Models in ML

Problem:

Using the Bank Data csv file, create a new notebook to train, test, and evaluate an ensemble model and compare the results with a single Decision Tree.

Abstract:

Cypress from New Ventures Department has found a banking client who wants to do a targeted marketing campaign for a specialty investment product to some of its customers. Cypress stresses that the bank only wants to spend the effort on customers who are likely to say yes; last year they used a different company to generate an algorithm and ended up spamming many of their customers with products they didn't want.

Data exploration.

• Data Cleaning
  • Looking for Missing Values
• Formatting the variables
  • Binary Variables
  • One Hot Encoding
• Visual Exploration of the variables
  • Categorical Variables
  • Numerical Variables

Analysis.

• Split Data Set
• Hyperparameter Settings
  • Choosing the best hyperparameters for the decision tree model
  • Choosing the best hyperparameters for the random forest model
• Comparison between Models
  • Visualize Confusion Matrix
    • Decision Tree Confusion Matrix
    • Random Forest Confusion Matrix
  • Visualization of the Trees
    • Decision Tree
    • Random Forest
• Performance Comparison

Summary/Conclusions.