Project Title: Cross Selling Prediction

For eight weeks, I had the privilege of working on a real-world application of Machine Learning: Cross-Sell Prediction. I first analysed the company's records and past data. Then, I used it to create an XGBoost classifier that predicts whether a customer would purchase 'Policy Y' if they already bought some 'Policy X'.

1. How does it work?

The cross-selling prediction system works by analysing customer data to predict additional products or services that a customer is likely to be interested in based on their past behaviour and interactions. Here's a general overview of how the system works:

2. About the repository and the internship:

1. Confidential data was used for the actual project and was provided to me by Iffco Tokio GIC. Hence, the working project can't be shared. I've uploaded some notebooks for reference. These have been stripped of their data and only give a glimpse into my work during the internship.
2. Housing.csv, boston.csv, and diabetes.csv were used as practice datasets. These were provided to me by Geshu sir. The housing dataset seems similar to this one.
3. Throughout the internship, I made notes and collected important links and topics. I used all that, along with the knowledge I gained while working on the project to create a comprehensive 'Project Report', which was submitted to Iffco Tokio GIC.
4. For more information, check out my LinkedIn post detailing my experience. You can even reach out to me for further queries and for collaboration!

I approached this valuable opportunity in this manner:

• took guidance from my mentors: Arpit sir, Geshu sir, and Sumit sir
  
• learned concepts from YouTube, Coursera, GeeksForGeeks, and Kaggle Notebooks
  
• made notes in my journal, text files, and Jupyter Notebooks (all of which can be found here in .pdf, .ipynb, and .txt files in here)

Note: This repository doesn't contain any information belonging to 'Iffco Tokio General Insurance Co. Ltd.'