Inspiration We were inspired by a desire to help underprivileged areas by creating a data-driven solution that could make a meaningful impact on community development. Our goal was to provide insights into tax certificate sales, a process that affects vulnerable neighborhoods, and help stakeholders make informed decisions.
What It Does The project moves from a broad neighborhood-level analysis to an individual property approach. We developed a machine learning model that assigns a probability to each property’s likelihood of being sold, enabling more precise targeting and decision-making.
How We Built It We used a combination of Python, R, Tableau, and Machine Learning algorithms. Python and R were used for data processing and model development, while Tableau was employed for visualization and presenting insights. Machine learning was applied to predict probabilities for each property sale.
Challenges We Ran Into Feature Engineering: Deciding which data points were most important for predicting sales was a challenge. Differing Ideas: Aligning on a unified approach when there were multiple ideas for solving the problem. Unclear Documentation: Navigating unclear or incomplete documentation for some of the tools we used made implementation more difficult. Accomplishments That We’re Proud Of One of our key accomplishments was successfully assigning probabilities to each potential sale. This gave stakeholders a better understanding of the likelihood of each property being sold, adding precision to decision-making.
What We Learned We gained deep insights into the world of tax certificate sales, and learned how to effectively use machine learning to predict outcomes and assign probabilities to each row in our dataset.
What's Next for Ospreys We plan to present and refine this project further. Our next steps involve improving the model’s accuracy, expanding the dataset, and working to make the tool more user-friendly for stakeholders in real estate and public administration.
Log in or sign up for Devpost to join the conversation.