Spam Classifier (2024):
I used my second AI coursework as an opportunity to explore one of my interests at the time, meta learning. I created a spam classifier with logistic regression, and used MAML for pretraining.
Sudoku Solver(2023):
My first AI coursework required me to create a sudoku solver that could solve 30 puzzles in 30 seconds or less. I applied Donald Knuth's algorithm X to these puzzles, ultimately solving the problems in 2.6 seconds.
Flappy Bird (2020):
One of my earliest projects was creating an agent that learns to play flappy bird in python using a feed forward neural network.
A.T.O.M (2020):
Agent Trained to Oppose Me (A.T.O.M) is a bot created to compete against the user in tic tac toe. It uses minimax to evaluate the board state and choose the best possible move.
EPQ (2021):
My Extended Project Qualification (EPQ) uses logistic regression to recomed films to a given user.
Summarizer (2021):
Uses text rank algorithm to estimate the importance of each line in a given text and returns the most important lines to give a general overview of what the text is about
Career Recommender (2022):
Uses an LSTM to perform setiment analysis on user's text input. K-Nearest Neighbour is performed on this data combined with data regarding other features (collected with unconvential methods) to recommend a career paths suited to the user.