Data Analyst

YANHENG LI

******@*******.*** 310-***-****

https://www.linkedin.com/in/yanheng-li-10a947ab/

EDUCATION

Cornell University, Ithaca Aug. 2016 - Dec. 2017

• Master of Engineering, Materials Science and Engineering Related Coursework: Application of Statistical Methods and Optimization (EXCEL), Linear Models with Matrices (R), Data Mining and Machine Learning (R), Statistics and Data Analysis for Physical Science (MATLAB), Project Management University of California, Los Angeles Sep. 2012 - June 2016

• Bachelor of Science, Materials Science and Engineering

• Honors: Dean’s List; Knapp Scholarship

• Technical Breadth area in Technology Management: Project Management, Marketing, Finance, Economics and Statistics SKILLS

• Programming: R, Python, SQL, MATLAB, Basic Java

• Statistics: probability, distribution, statistical inference, hypothesis testing, Bayes theorem, ANOVA

• Machine Learning: supervised learning (linear regression, logistic regression, decision tree), unsupervised learning

(clustering, optimization, model evaluation, recommendation system)

• Data Visualization: histograms, frequency polygons, box-plots, quartiles, scatter plots, heat maps, ROC

• Tools: Excel, MATLAB, MySQL, Hive, Presto, Tableau, PowerPoint, Word, Outlook WORK EXPERIENCE

Data Analyst Intern, Ele.me, Shanghai, China May 2017 - Aug. 2017

• Extracted data through MySQL and created real-time food delivery dashboard using Eleme Multi-Dimensional Analytics and Tableau to visualize daily 8M orders in heat map, and monitor market performance based on selected time frame

• Improved current commission and subsidy rate of brand restaurants through forecasting sales performance, loss evaluation, fans growth, customers quality and substitute cost using least-squares trend estimation which reduced $7M cost in 1 month

• Analyzed competitors’ operational performance based on 23 key features with billions of data and collaborated with Business Development department to conduct new operational strategies resulting in 12% increase in daily GMV

• Designed selective user analysis function with engineer team on current data product Treasure Map to help restaurants understand their customers and recommend marketing strategies which reduced advertising cost by 21% and increased daily sales by 15%

Process Engineer Intern, Corning Inc., Corning, New York Jan. 2017 - May 2017

• Revised current glass batch pelletizing process to develop tailored procedure to fit product needs under our lab conditions

• Defined the quality of batch pellets by their particle sizes and strength, and built a Linear Regression model to make predictions based on features, such as materials selections, water percentage, binder percentage, rotor speed and pan speed

• Developed a new pelletizing process based on predictions, which yielded an increase in pellets quality by 20%

• Conducted a scientific report and presented it to the lead of the Materials Department Business Data Analyst Intern, Bailian Group, Shanghai, China June 2016 - Aug. 2016

• Conducted event planning through both Flash Sale and Omni Channel

• Developed predictive models to forecast market size and recommended new operation strategies

• Created a performance-based dashboard using Tableau and R for C-levels to improve data accessibility and interpretability

• Classified Bailian members into different groups and developed targeted-user marketing strategies which help company provide customized services and benefits based on users' behaviors PROJECT EXPERIENCE

Predicting Housing Prices in Ames, Iowa Jan. 2017 - Mar. 2017

• Extracted 79 features from raw Ames housing data containing different types, such as categorical, numerical and time series data, and imputed missing data using chained equation (MICE) algorithm in RStudio

• Performed feature selection through exploratory analysis including visualization and hypothesis testing

• Fitted linear regression model with regularization to control for multicollinearity

• Built decision tree, random forest and boosting decision tree to predict housing price and achieved 0.096 RMSE by boosting decision tree model on test data set

Contact this candidate