Tong He

EDUCATION

Sun Yat-Sen(Zhongshan) University

B.S. in Statistics, Sep. 2009 - Jun. 2013
- Overall GPA: 3.6/4.0
- Last two years GPA: 3.9/4.0

Augmenting LASSO Regression with Decision Tree for Identifying the Correlation of Genetic Polymorphism and Adverse Events
- Accepted as regular paper by BIBM 2013
Introductory Statistics with R in Chinese
- One of the translators of the Chinese version, originated by Peter Dalgaard

Single feature polymorphisms gene detection based on Bayes model
- The basic idea for this project is dealing with small samples, modelling with hierachical Bayes model, and computing result via EM algorithm
- Undergraduate thesis, rated as the top level.
Microarray Data with shrinked MANOVA statistic, analysed based on FDR
- Combining MANOVA statistics and shrinked statistic to accurately estimate false discovery rate
- Borrow technics from Large Scale Inference by Bradley Efron

Douban
Algorithm Engineer Intern, Beijing, China, 2013
- Douban.com is a famous SNS website in China with the largest database of Music, Books, Movies, etc.
- Predict user gender and income level via preference data on books, movies and groups, with the final precision higher than 95%
- Clean and structurize folksonomy data of Movies, the procedure becomes the standard in Douban
- Natural language processing involving Chinese word splitting, and construct an internal word dictionary of Douban
Giordano
Buisness Analyst Intern, Guangzhou, China, 2013
- Mine influencial factors associated with sales trends
- For instance, how is season, temperature, city, weather and holiday influence the performance of stores in an area.
Industrial and Commercial Bank of China
Data Analyst Intern, Guangzhou, China, 2012
- Organize and formalize financial data for enterprises
At Home Academy
AP test Teacher, Guangzhou, China, 2011
- Teach AP Computer and AP Statistics to high school students, explain basic topics in statistics and java
- Assign homeworks and projects related to real world problems, help them prepare for the AP test
- All students got at least 4/5 in the final test

Cleansing and analysis of folksonomy in Douban
- A presentation on my analysis procedure, which was developed as an intern in douban, of folksonomy data
- Read it Online : Sixth useR China conference in Shanghai, Nov. 2013
Ensemble model in machine learning
- An introductory presentation on Caruana et al. Ensemble Selection from Libraries of Models and its variation
- First R Salon in Guangzhou, Jun. 2013

KDD CUP, 2013
- Top 20% in track 1
ACM/ICPC, 2012
- Third prize in Guangdong province
- Second prize in SYSU campus
Individual Study Scholarship, 2012
- Dedicate to those excellent in research and study
National Mathematical Modeling Competition, 2011
- Third prize in Guangdong province

R
- Seasoned R user
- A package rLTP as the interface between R and the ltp cloud service which is a tool for Natural Language Processing in Chinese.
- Having a project R for Codejam aiming to implement classical algorithms in R
C/C++
- Experience with ACM/ICPC in college and Olympiad in Informatics in high school