Attrition rate is one of the most important factor which determines the performance of the product. It gives us an insight into how many people are leaving the services. Once this information is obtained, the company can take some actions to modify their SOPs to cater to its customers. In this project, 3 machine learning algorithms, namely Support Vector Machine, K Nearest Neighbors and Logistic Regression have been employed, to determine the likelihood of a customer leaving or staying with the Credit Card services.
streamlit-main-2021-03-21-23-03-01.mp4
The Dataset has been downloaded from Kaggle. Check out the link or directly download it from here
Project is created with:
- Python: 3.7.10
- Streamlit version: 0.77.0
- Heroku version: 20
Please check the requirements.txt file to know about more packages used in this project.
The problem is primarily a binary classification case where the algorithms like SVM, KNN and Logistic Regression can perform a good task. Each algorithm has its own advantages and disadvantages and thus a better insight can be drawn when all the 3 classifiers are applied to the one data point under consideration.