Human Activity Recognition
Performed various Deep Learning techniques: DNN, LSTM, Conv-LSTM, … to detect human activities generated by sensor-based wearable devices
About me
MSc. in Analytics at UChicago
Background in Market Research
Hi, I’m Tam. I grew up in Vietnam and did my undergraduate in Economics at Toulouse I Capitole Université.
With a quantitative background, I started my career as a Market Researcher, specialized in Asian Market.
Over the past six years, I had many chances to work and gain knowledge in various sectors across different types of projects, including but not limited to Segmentation,
Brand Heath Check, U&A, Tracking, Concept - Product test, Pricing, and Consumer Satisfaction.
Such opportunities also paved the way for me to realize how passionate I am about uncovering the hidden insights out of the data. My long-term objective is to utilize my knowledge and skills to help improve people’s lives;
however, traditional research has its limitations. Therefore, I made a call to step out of my comfort zone, starting a journey of thousand miles to the US to obtain a rigorous and robust knowledge in Analytics & Data Science at UChicago.
After graduating in Summer 2020, I started my new journey as a Data Scientist at Boehringer Ingelheim – one of the world's 20 leading pharmaceutical companies, and the largest private one.
I am also opening to collaborate on side projects related to Machine Learning, Deep Learning, NLP, and Healthcare. You can reach out to me either via email
or Linkedin. I am happy to connect! :-)
Fun Facts
Projects
Below are some of the Analytics Projects that I had worked at school
ICU Survival Analysis
Using data within first 24 hours of intensive care to develop a machine learning model that could improve the current patient survival probability prediction system (apache_4a) and is more generalized to patients outside of the US
Big Data Solution for Gaming eCommerce Platform
Applying Big Data Platform and Technique to: 1) Understand gamer behavior and habits. 2) Predict user playtime and build recommender using tree based regressors, classifiers. 3) Group players with similar attributes, and make recommendations using ALS and Graph Algorithms
Soccer Scouting
For this project, we are positioning ourself as a scouting agency that uses analytics to, among other things, enhance the discovery of talents and help soccer clubs better understand the dynamics (features) that come into play when determining the value, overall and future potential of a player
Drought Forecasting - Time Series
Identified the most appropriate Time-Series method to forecast drought in African countries, acting as a critical early warning for drought managements
Divvy - Chicago's Bike Sharing System
Developed a relational database that will enable quick response and analysis on the current state of Divvy’s operations in regard to ridership, station locations, other factors affecting them. Then built a scoring model to optimize the number of stations and bikes allocated by zip codes
