These projects highlight a range of experiences in applied machine learning and data analysis. The work includes developing a stock market predictor using CNNs, building a movie recommender system with the Apriori algorithm, and creating a predictive model for sorting user feedback on Twitch. Additionally, a capstone project involved leading a team to classify floor plans for Steelcase, culminating in a practical tool adopted by the company. Each project showcased skills in team leadership, data handling, and applying advanced algorithms, emphasizing learning through hands-on challenges and collaboration.
SteelCase Capstone
Led a team to develop a binary classifier for assessing the suitability of floor plans for training models. The project included being the main contact with the client, organizing team efforts, and meeting strict deadlines. Additionally, a Streamlit application using K-Nearest Neighbors (KNN) was developed, leading to new implementation plans by the Steelcase team.
CSE 881 - Data Mining: User Feedback Sorting on Twitch
Collaborated with two other students to create a predictive model that categorized user feedback on Twitch. The project involved web scraping 13,000 data points and data cleaning using Python. The team developed the model through regular meetings and brainstorming sessions.
CSE 482 - Big Data Analysis: Movie Recommender
In this team project, we developed a movie recommendation system using the Apriori algorithm on a dataset of over 25 million data points. My role involved collaborating with four other students, setting checkpoints, integrating the algorithm, and creating a Streamlit interface in Python to demonstrate the algorithm's effectiveness in real-time.
CMSE 890 - Applied Machine Learning: Stock Market Predictor
A solo project aimed at predicting stock market trends using Convolutional Neural Networks (CNNs) with candlestick charts. Despite a steep learning curve and achieving only 51% accuracy, the project was a valuable learning experience in building a CNN from scratch. The takeaway was the importance of leveraging existing, optimized solutions.