I am a

Data Scientist

I am interested in leveraging analytical tools to make better decisions. As a data scientist, I am passionate about learning algorithms and methods and figuring out how these help us make decisions.

Statistician

With a solid background in theoretical statistics, I perform quantitive and qualitative analyses for the study development.

Data Analyst

I have hands-on experience in data processing including data collection, data manipulation and data visualization etc.

Consultant

I have worked for a consulting start-up where I have learned how to pitch myself to clients and elaborate ideas to the non-tech people.

Featured Project

A Neural Image Caption Generator

An end-to-end neural network system that can automatically view an image and generate a reasonable description in plain English.
[Deep Learning, NLP, CNN, LSTM, Computer Vision]

Elo Merchant Category Recommendation

Processing the transaction information and design a model to predict the customer loyalty.
[LightGBM, Feature Engineer, Customer Loyalty, Cross Validation]

Sentiment Analysis for IMDb Movie Reviews


An architecture based on RNN-LSTM to predict sentiment for movie reviews from IMDb.
[Deep Learning, NLP, RNN, Sentiment Analysis, Data Mining]

Pollinator-Habitat-Restoration-Tools

Statistical consulting project on Pollinator Habitat Restoration Tools. Designing a R-shiny app for visualizing the laboratory dataset.
[Statistical consulting, ANOVA, R-shiny, Data Visualization, Experiment Design]

Isolation Forest

Following the original papers, reproduce the anomaly detection algorithm from scratch, with improvement on noise resistance.
[Python, Anormaly Detecting, Object-oriented Programming]

Variance component analysis for proportion data using zero-inflated beta mixed-effects model

I propose an ZIB model based approach for variance component analysis for zero inflated proportion data. Simulation studies demonstrate that our approach is more robust to the excess number of zeros than GLMMs, and can give good estimate when GLMMs fail to converge.
[Variance Component Analysis, Simulation Study, Zero-Inflated Model]

Spark Music Recommendation System(On-going)

Using Spark, Design a music recommendation system
[Spark, Python Flask, Recommendation System, ALS Algorithm]