Arjun Lal
I recently graduated from the University of Pennsylvania, where I earned both an MSE in Computer Science and a BSE in Computer Science with minors in Statistics and Data Science. While at Penn, I held several graduate and undergraduate teaching assistant positions in classes ranging from Algorithms to Machine Learning.
I spent summer 2019 interning as a Data Scientist at Facebook. My work focused on investigating the advertising vs. user-engagement trade-off on Stories and I ended up finding ways to optimize for revenue with minimal impact on user-engagement. I will return full-time August 2020.
I'm interested in data, product strategy, software development, and investing. Feel free to check out the links below and get in touch!
Recent Projects
Event-driven exploration of airfare prices and routes
Work from the Citadel East Coast Invitational Datathon. My teammates and I used a Latent Dirichlet Allocation Model and regression analysis to see how event variety in major US cities affects airfare prices. We ended up winning 1st place and qualifying for the national championship!
Removing bias in sentiment analysis systems
Final project for CIS 530 (Computational Linguistics). The goal of this project was to build a linguistic and statistical model that could predict the relative intensity of anger in a given text with minimized gender bias, in an effort to decrease the algorithmic bias that pervades current systems.
Determining important application factors in university admissions
Paper for STAT 474 (Modern Regression). Given data describing applications to a hypothetical university, I used a Generalized Additive Model to determine which factors are most relevant in deciding admission and how well they can predict admission.
Personal website
I designed and built this website to get myself comfortable with HTML, CSS, jQuery, and Bootstrap. I hope you can get a look at what I've worked on and what I'm interested in!
Query-based taxi service
Final project for CIS 550 (Database and Information Systems). Using early 2019 NYC taxi rides, we built a dynamic and query-based platform for both drivers and riders that would help them plan their respective routes as efficiently as possible.
PennBook
Final project for NETS 212 (Scalable and Cloud Computing). We built a mini-social network with friending, chat, profiles, etc. In addition, we used MapReduce to generate friend recommendations based on common friends and affiliations via the adsorption algorithm.
Books And Articles I've Liked
