Data Science Projects
Board Game Cafe Database
As part of a four-person team, I helped create a relational database to support the operations of a board game café. I took the lead on designing the database structure, creating the ER diagram and schema that defined six interconnected entities with multiple attributes. This design ensured that the system could manage games, customers, and transactions efficiently. I also developed SQL queries and stored procedures, which were integrated into a front-end website to make the database fully interactive.
Taskly - Task Tracker Iphone App (2025)
I developed a mobile task management app in Unity with an intuitive, easy-to-use interface and a scalable date-based task system. The app supports an infinite sequence of days, seamlessly loading saved tasks as users swipe between them. Features include smooth scrolling, context menus, task editing, and a “move to next day” option for streamlined organization. To ensure reliability, I implemented a JSON-based saving and loading system for persistent use across sessions. I am also designing expanded settings and calender functionality while preparing the app for submission to the iOS App Store.
Kindle Data Analysis
I worked with a Kindle dataset of over 100,000 entries, focusing on cleaning, processing, and extracting insights from the data. Using techniques such as Linear Regression, Principal Component Analysis (PCA), K-Means clustering, and Gaussian Mixture Models, I explored relationships within the dataset and predicted book ratings and performance based on features like category, author, and reviews.
AI-Driven Public Transportation Routing
Collaborated in a six-person research team to simulate traffic flow and explore ways to improve efficiency using AI algorithms. Together, we implemented graph structures and Dijkstra’s algorithm to route buses along optimal paths and proposed demand tracking with demand-based routing as a strategy for further optimization. Our work was presented to judges in a competition against other research groups.
Chess Bot With Variable Computational Loads
Developed a fully functional chess game and AI bot in the Java terminal. The bot’s AI is built using probability mapping to evaluate potential moves, incorporating a configurable lookahead of multiple turns to simulate strategic planning. Users can adjust the difficulty by controlling the number of turns the bot calculates and the level of randomness in its decisions, allowing a balance between optimal play and unpredictability. The system demonstrates core AI principles, including decision trees, heuristic evaluation, and probabilistic reasoning.