Inspiration
Everyday, thousands of people attempt to make financial decisions with lengthy processes, one of them being home loan approval. With this project, we seek to make the process more accessible by bringing this process one more step closer to automation, that is, allowing people to calculate whether or not they are eligible for a loan. In addition, the models created in this project provide an insight to how an applicant's background effects their likelihood of being approved for a home loan.
What it does
This project takes a set of data which contains the background information of applicants and learns from this data to make predictions on whether or not a given applicant should be approved for a loan.
How we built it
We used google colab run the machine learning models. Some of the models we built include logistic regression, naive bayes, and support vector machines. We used various python modules and libraries to aid this process such as numpy, pandas, sklearn, seaborn, matplotlib, amongst others.
Challenges we ran into
One challenge we ran into was the fact that our data set was not perfect right of the bat and required pre processing. For example, some of the fields in certain rows were empty and some of the data was categorical and not numerical.
Accomplishments that we're proud of
We're proud of the fact that we were able to create multiple prediction models with very little knowledge of machine learning.
What we learned
While researching ways to go about this project, we learned briefly about various topics within machine learning such as supervised vs unsupervised learning, and about different machine learning models including linear regression and neural networks, although we did not utilize all of these.
What's next for Home Loan Approval Analysis
Hopefully this analysis brings us one step closed to creating a model that takes any arbitrary applicant's information and predict whether they are approved or not.
Log in or sign up for Devpost to join the conversation.