Inspiration
There's a lot of factors that affect customer retention. For larger companies, the dataset might be too big and the correlation between factors are too complicated for us to draw straightforward conclusions. We wanted to simplify this problem, and we thought data science and machine learning are excellent tools to aid us in this process.
What it does
Our model look at churning of customers at any company - A telecom company, in our example. We manipulated the data and use python libraries for data visualization. After cleaning the data, we then used machine learning algorithms - Linear Regression and Suppor Vector Machine - to recognize determining factors in customer churning. The correlations between each factor and customer churn rate becomes more obvious, and this insight will help the company develop targets in strategies to improve customer churn rate.
How we built it
Our project is built on Jupyter notebook using python. We used the machine learning models in python's scikit-learn library. Our project also includes multiple graphs to support our explanation.
Challenges we ran into
Some challenges we ran into include finding a good topic to work on, and we lost 2 teammates because of this. It was hard to find an interesting topic in retail while making sure that there are datasets online that fits our needs.
Accomplishments that we're proud of
This is our first time using the scikit-learn library and we are proud of what we have accomplished. We got to turn our theoretical knowledge on machine learning into practical work, and give meaning to our data.
What we learned
We learned how to make graphs in python, became efficient in using scikit-learn, manipulated data and drew conclusions.
What's next for BabyDon'tGo
Model from BabyDon'tGo can be applied to other companies to gain insights into churn rate, or the machine learning concepts can be applied to other categories to determine target area as well.
Log in or sign up for Devpost to join the conversation.