Stuck Silver Heirs Property Solution

Inspiration

Helping those in low-income neighborhoods in Jacksonville receive support or financial relief so they can pay off the tax credit on a house they inherited from a loved one that passed away. We mainly focused on finding any discrepancies between people of different ages and sexes based on the average appraised value of the homes in the neighborhoods.

What it does

Based on the location, age, sex, base bid, assessed value, and several other factors/variables. We were able to create a couple of models that could assess whether a person was more likely to have their house redeemed or sold by imputing the variables above.

How we built it

We used Python, scikit-learn, pandas, and Microsoft Excel. We created a logistic regression, random forest, and gradient boost model. They used the independent variables of location, age, sex, base bid, assessed value, and several other variables to predict our dependent variable named "Code" which tells us whether a person paid off their tax credit and redeemed their house or if it was sold.

Challenges we ran into

Data cleaning and pre-processing, handling nulls in our dataset.

Accomplishments that we're proud of

We presented one of our more successful models.

What we learned

How to handle messy datasets and use ML to help solve real-world problems.

What's next for Stuck Silver Heirs Property Solution

Testing other factors that may go into a person who paid off their tax credit and those who didn't. Looking into race, level of income, and single parent households may help us identify who else may need help in paying off their tax credit and sending them financial relief.

Built With

• excel
• pandas
• python
• scikit-learn