Skip to content

An insurance company has a historical data set (train.csv). The company has also provided a list of potential customers to whom to market (test.csv). From this list of potential customers, the model determines whom to market and whom not to.

Notifications You must be signed in to change notification settings

Hari-Dorbala/Propensify

Repository files navigation

Propensify

About:

An insurance company has provided a historical data set (train.csv). The company has provided with a list of potential customers to whom to market (test.csv). From this list of potential customers, the ML model determines whom to market to and whom not to and saves the same to an Excel file (test_with.predictions excel file).

Given Files:

train, test

Generated files

A. Documentation_Capstone_Propensify.pdf : It explains the reasoning behind various steps involved in the model training.

Contents present in Documentation_Capstone_Propensify:

  1. Introduction
  2. Treating Missing Values
  3. Feature Engineering
  4. One hot encoding categoric features and normalizing continuous features
  5. Choice of sampling
  6. Choice of metrics and model
  7. Model
  8. Utility of the Model

B. SourceCode_Pipeline_Capstone : It contains the pipeline of operations. It generates the required file, i.e., test data with predictions on whom to market and whom not to.

C. test_with.predictions : Excel file with predictions

D. Documentation_Capstone_Propensify.ipynb : Contains the ipynb file of Documentation_Capstone_Propensify.pdf document.

E. propensify_model.joblib : Contains joblib file of the trained model created and used in SourceCode_Pipeline_Capstone

F. preprocessing_pipeline.joblib : Contains joblib file of preprocessing pipeline created and used in SourceCode_Pipeline_Capstone

NOTE: Please change the path to files to run the source code. The current path is set to the path at which files are present in my system. Please run the SourceCode_Pipeline_Capstone ipynb file to get the predictions.

About

An insurance company has a historical data set (train.csv). The company has also provided a list of potential customers to whom to market (test.csv). From this list of potential customers, the model determines whom to market and whom not to.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published