Fulcrum Labs Internship

I am currently a Data Science intern for Fulcrum Labs, a start up developing an adaptive learning platform. During my first quarter at the company I devised a natural language processing (NLP) algorithm that will determine if a student’s response to a question is related to a body of text they were to read. Furthermore, I prototyped an algorithm that will assign a performance score to a student given a report from an instructor about a student’s behavior in class. This performance score and be augmented by how well they complete courses on the platform.

I am currently working on a prediction model that will, based on how a student chooses to do with their time on the Fulcrum Platform, determine how much time it will take for as student to complete a course. Using keras, I implemented a Long Short Term Memory (LSTM) to predict how much time a student will spend on a course.

So far, I have trained it on the data from one student, and I am working on having it train for multiple students to improve predictions.

This internship required a lot of independent research and work. During my research I have played around with various algorithms for gaining insight from data such as Support Vector Machines (SVM) and Google’s Page Rank algorithm. Along the way, I learned how to clean up and structure my data in order to implement these algorithms.