adudqs@r.postjobfree.com LinkedIn: Zixuan Liu 443-***-**** Baltimore, MD

EDUCATION

Johns Hopkins University Baltimore,MD

MS in Engineering Management GPA: 3.6 May 2023

Beijing University of Chemical Technology Beijing,China BS in Macromolecular Materials & Engineering GPA:3.76 July 2020 Relevant Coursework: Deep learning, Machine Learning, Computational Applied Mathematics, Probability and Statistic SPECIALIZED SKILLS

Data Analytics: A/B Testing, Casual Inference, Tableau, PowerBI, SAS, Google Analytics, MS Excel

Machine Learning: Python (Scikit-Learn), R, SPSS Predictive Modeling, Hypotheses testing, Regression Analysis

Database Processing: SQL, AWS S3, Google Cloud BigQuery Data Engineering: Spark, Hadoop, Hive, MapReduce

WORK EXPERIENCE

JD.COM [Online Shopping Platform] Beijing,China

Data scientist Intern February 2021-Auguest 2021

Engineered 161k customer data from database and extracted and engineered 15+ usable features including age, city

Identified the key customer churn point following AARRR model for 5 cohorts and compared the difference

Analyzed the impact of promotion activities in a weekly basis of sales data in order to better optimize the strategy

Conducted A/B test for two different promo types and implemented the statistical methods to analyze the results TAL [Online Shopping Platform] Beijing,China

Customer Analyst Intern July 2020-October 2020

Created 200+ questionnaires related to the customer experience and collected all the information in two weeks

Identified the customer demands based on the feedback of customers and analyzed the user responses

Predicted the customer churn rate in linear regression model and XGBoost, reached R2= 0.87

Conducted the competitor analysis for 20+ different competitor following PEST and SWOT framework

Optimized the products based on the market analysis, improved the revenue for 20% ACADEMIC EXPERIENCE

Natural Language Processing and Topic Modeling August 2021-October 2021

Conducted sentiment analysis on 1M+ comments using NLP, achieved out-of-sample accuracy 85.7%

Preprocessed review text by tokenization, stemming, removing stop words and extracted features by Team Frequency-Inverse Document Frequency(TFIDF).

Performed unsupervised learning models of K-means and Latent Dirichlet Analysis to better cluster the important topics related to certain products.

Visualized model training results by dimensionality reduction using Principle Component Analysis. Auto-Encoder-Decoder and embedding framework in movie Recommendation August -October 2021

Performed analysis on IMDB movie data and built an auto-encoder-decoder in Python to predict the ratings a user would give a movie.

Preprocessed data by removing missing values, data exploration and splitting data into train and test datasets.

Prepared user-movie-matrix based on users ratings and built and trained a deep learning model with 3 hidden layers.

Evaluated the model based on its performance(RMSE) on test datasets and tuned the hyper parameters to achieve a better prediction accuracy.

Customer Churn Prediction Project January 2020-February 2020

Developed algorithms for telecommunications service vendors to predict customer churn probability based on labeled data via python programming and Apache Spark.

Preprocessed data set by data cleaning,categorical feature transformation and standardization, etc.

Identified 10+ important features related to promote customer churn in Logistic Regression, Random Forest and K-Nearest Neighbors, and applied regularization model, achieved AUC=0.89

Improved prediction performance by 1% (measured by AUC) via with 5-fold cross-validation and GridSearchCV

Contact this candidate