A very special thanks to those who helped shape my experience throughout my time at Metis.
Metis is an immersive data science bootcamp in San Francisco, CA with a curriculum designed by Datascope Analytics. Through five data projects, Metis teaches and reinforces concepts such as statistical inference, data acquisition, supervised and unsupervised learning, D3.js, SQL and NoSQL databases, and Hadoop MapReduce.
Data Science Projects
RecipEat: Recommend Recipe from Image of Ingredients
• Built a recommender application that suggests recipes from an image of ingredients.
• Trained a multi-label image classification model using convolutional neural networks.
• Used Keras as neural networks API with Tensorflow backend for deep learning.
• Acquired images web scraping with Selenium, and stored data using AWS.
• Used Flask as web framework and Pure CSS to design web app.
Political Party Affiliation Tweet Predictor
• Built a classification model to predict whether a tweet is more republican or democratic.
• Applied NLP techniques on user timelines with an established political party affiliation.
• Acquired tweets through the Twitter REST API, and stored data using MongoDB.
Classifying Marijuana Abusers with NSDUH
• Created classification model to predict if an individual suffers from (marijuana) addiction.
• Used National Survey on Drug Use and Health (NSDUH) to engineer features ranging from age of first usage to cross addiction with alcohol.
Predicting Fantasy Football Points
• Made regression model with players' previous years stats to predict fantasy football points.
NYC Turnstile Data
• Cleaned and analyzed NYC turnstile data to map out the ridership profile of a subway station.