SUBSCRIPTION PROBABILITY MODEL

Using machine learning to develop models that is capable of predicting customer subscription based on current dataset

The dataset is composed of 21 variables: [20 independent (X) & 1 dependent(y)]

• The code analyses the spread of the current dataset
• Calculates the correlation between the input datasets and the output dataset
• Identifies 5 most correlated x(i) variable to y
• Model 1 - Linear regression model using 20 inputs
• Model 2 - Linear regression model using 5 most correlated input to 'y'
• Splits the data set to Training and Test Data set with a ratio of 9:1
• Evaluates the accuracy of the machine models

PREREQUISITIE:

1. The program is scripted in Python3. Have the correct version of python installed
2. Please ensure you have following libraries installed:

• pandas
• numpy
• matplotlib
• scipy
• sklearn

3. Please have the following files in the same directory as main.py

• data set.csv

INSTRUCTIONS:

1. Load Terminal
2. Set current directory to the folder containing the main.py file.
3. Run script from a Terminal. Type python3 main.py
4. The plots generated by the code will be saved in the same directory as main.py file.

Author

• Jerin Philips Rajan