In this project, I used social network analysis techniques to identify key members of a collaboration network from an innovation contest.
In this second look at innovation collaboration networks, I evaluated using a K-means clustering model fit to various network centrality measures to identify influential innovators. This was my class project for Stanford's graduate class Social and Information Network Analysis CS224W.
In this project, I developed and evaluated a new Network Guided Naive Bayes (NGNB) classifier that uses undirected graphs to accurately and quickly predict labels for multi-topic text. This was my class project for Stanford's graduate class Machine Learning (CS229).
Solution for the Kaggle Amazon Employee Access Challenge. The goal was to create a prediction model to approve or reject access requests based on an employees organization, title, etc.
Solution for the Kaggle Packing Santa's Sleigh challenge. The goal was to pack one million presents of various sizes in the smallest volume possible while optimizing for delivery order.
Solution to the Kaggle TensorFlow Speech Recognition Challenge. The competition’s goal was to train a model to recognize ten simple spoken words using Google’s speech command data set. This is essentially the trigger word detection problem that alerts voice activated intelligent personal assistants when to pay attention.
Solution for the Kaggle Toxic Comments challenge. The goals was to train a model to detect toxic comments from Wikipedia edit pages.