Built at IvyHacks 2020. Allow users to submit a YouTube video or text block and return a text summary. Built with React JS on front end and summary algorithms are built using machine learning and text-rank algorithms using Python and Tensorflow. These algorithms are hosted using Python Flask and Heroku.
Web API based off my previously built KPopy Python module. Allows users to gather information about Korean artists using GET requests. Hosted using Heroku and Python Flask.
I built a multi-layer perceptron classifier and regressor to answer what distinguishes between a top K-Pop group such as BTS and Twice and the other lesser known artists. Was selected by Medium's curator and featured on other data science websites.
I gathered thousands of Donald Trump's tweets and League of Legends subreddit comments using tweepy and PRAW. Using Max Woolf's text generating recurrent neural network, I let the neural network train over dataset for over two hours to see what the neural network comes up with when it generates its own text.
Python module that allows users to gather thousands of information about Korean artists such as their stage name in English and Korean, real name, birthday and place, the group they're in, and many more.
My 2nd semester project for my Independent Studies for Computer Science class in my senior year of high school. Used techniques such as Ngrams, TF-IDF vectorizer, and Bayesian Statistics (Naive Bayes Classification) to train over a data set that classifies thousands of words and phrases with a sentiment. Input any URL to get the average sentiment of that site.
Gather hundreds of data from TypeRacer.com and analyze it. Focus on looking to see if there is any correlation between WPM and the number of words per passage and other various potential factors. Discuss the plausible effects when not conducting a proper experiment.