Machine Learning Projects & Algorithms
A comprehensive collection of machine learning algorithms and projects, from decision trees to neural networks, implemented from scratch and applied to real-world datasets.
Featured Projects
Conversational AI Pipeline
Advanced multimodal AI system that can see, hear, and understand context through real-time processing of audio, video, and screen content for intelligent conversational interactions.
AI Hosting Application
A full-stack platform for generating, managing, and hosting AI-generated content with authentication, dashboard, templates, and billing. See the full project on GitHub: https://github.com/mfg5169/AI-Hosting-Application
Financial Anomaly Detection
Comprehensive AWS-powered financial anomaly detection system with LSTM neural networks, real-time alerts, and LLM integration for stock market analysis.
Reinforcement Learning Algorithms
Developed reinforcement learning algorithms to play various games, including slot machines and other interactive environment games.
Neural Networks on MNIST
Implemented Neural Network models on the MNIST dataset to analyze the effects of batch normalization and compared the performance of MLP vs convolutional networks.
RNN vs LSTM Sequence Analysis
Ran anaylsis comparing the accuracy of RNNs and LSTMs in identifying sequence addition patterns and analyzing their performance differences.
Naive Bayes EM Algorithm
Developed a Naive Bayes Expectation-Maximization algorithm trained on over 100 documents of speeches, from 1964 and after, to predict political party affiliation.
Decision Tree Algorithm
Developed a machine learning decision tree algorithm to predict if a student will attend an ivy league school.
K-Nearest Neighbors Regression
Developed a k-nearest neighbors regression algorithm to predict if an individual will attend an ivy league school.
Source Code
All project source code is available on GitHub, where I'm continuously working on improvements and new implementations.
View All Projects on GitHubInterested in Collaborating?
If you'd like to learn more about these projects or discuss potential collaborations, I'd love to hear from you!