Mae-Proval

• 

  Analyzing Correlation of Customer Data Variables in Heatmap

• 

  Modified Customer Data Table

Inspiration

Demystify mortgage financing for clients

What it does

Our app asses whether a customer is fit to buy a home or not based on their expenses and income. The data of the customer can be obtained through the user inputting each individual expense and their monthly income, or from pulling the customer's data from a database with all customers' data. The app decides whether a potential homebuyer is deemed ready to buy a home or not by calculating their Loan-to-Value (LTV), Debt-to-Income (DTI), and FEDTI (Front-End-Debt-to-Income), and validating whether these values, as well as their credit score, meet the standards of a customer financially stable enough to own a home. If certain values don't meet the criteria required, the app gives suggestions on how to improve these values.

How we built it

We coded in Android Studio primarily using Kotlin, Jupyter Notebook, and MATLAB.

Challenges we ran into

Throughout this process we found it difficult to parse through the CSV file and sort the respective values into their approval rating variables. For the backend development, it was also difficult to navigate through the various variables of a heatmap.

Accomplishments that we're proud of

We were able to create a functional prototype within Android Studio and its user interface model. We were also able to create visualizations in MATLAB that showcased our data manipulation and created a full implementation of our app's functionality in Python through Jupyter Notebook.

What we learned

We gained insight on how to work the constraint layouts in Android Studio and how to overall navigate Jupyter Notebook.

What's next for Mae-Proval

We hope to implement more specific metric improvement suggestions based off of user input and additionally we would like to utilize a CSV library for easier data parsing.

Built With

• android-studio
• git
• jupyter
• kotlin
• matlab
• python