Portfolio - brentthomasmartin.com

Data Driven Solutions to Shape the Future

Explore innovative projects that leverage advanced data science, machine learning, and predictive modeling to unlock insights and power intelligent decision-making across industries. Each project page contains links to dashboards and Jupyter notebooks for deeper exploration and hands-on experience.

September 2024

Stock Prediction with Machine Learning Study

Building Predictive Models for Financial Markets

This study focuses on predicting stock price movements using machine learning algorithms, with a special focus on FAANG stocks and the S&P 500 index. Various models, including Random Forests, Gradient Boosting, and Long Short-Term Memory (LSTM) networks, were employed to predict stock trends with a strong focus on achieving high directional accuracy. The results demonstrated a robust 57% directionality prediction for FAANG stocks and a near 90% win rate for S&P 500 predictions.

August 2024

Federal Reserve Economic Data Study

Analyzing Trends and Insights from U.S. Economic Data

This project is a deep dive into U.S. economic indicators using the Federal Reserve Economic Data (FRED) API. The study identifies significant trends in employment, wage growth, GDP, and inflation using advanced data analysis techniques. Visualizations were developed to showcase the relationships between key economic metrics, with a focus on employment data and national economic health.

Coming Soon:

Forensic Analysis of the Enron Corpus using Natural Language Processing

Applying advanced machine learning techniques to uncover hidden patterns and insights from the famous Enron email dataset. This project will focus on fraud detection, sentiment analysis, and communication network analysis.

Image Generative Audio and Fourier Transform Visualization

Exploring the intersection of visual art and sound by generating music from images using deep learning models and creating visual representations of audio through spectrograms. This project will delve into creative AI applications in multimedia.

Interactive Game: NBA Fantasy League

Experience a dual-format NBA fantasy league, featuring both a conventional league for current players and a unique ‘Legacy League’ where you can draft and compete with retired legends. In the Legacy League, player performance is simulated through random draws of historical games, adding an element of nostalgia and unpredictability. Using advanced data analytics and game simulations, this project will provide an engaging platform for fantasy enthusiasts to explore past and present NBA performances in a competitive fantasy format.

Exploring Cloud-Based RDBMS and Data Integration for Scalable Analytics

I’m currently leveraging a robust RDBMS integrated with Google Cloud Console and BigQuery for large-scale data processing. This system enables seamless connections with Tableau and IBM Cognos for real-time analytics and dashboarding, while automating ETL workflows for efficient and scalable data pipeline management. Ideal for managing data across industries and ensuring smooth reporting and analysis.