Computer Science Web Developer

Michelle(Huayu) Zhou

Address: **** ******** **, *** **** 95133 Cell: 775-***-****

Email: ******************@*****.***

SKILLS

Languages: Java, C/C++, Python, HTML5/JavaScript, CSS3, JSON Data Science: mathematics, statistics and probability, classification, clustering, deep learning, optimization methods, supervised and unsupervised machine learning algorithms, natural language processing

Platforms: Apache Tomcat, MAMP, Eclipse, JUnit, JMeter, Android Studio, MySQL, MongoDB, Git/Github, Amazon EC2, Gradle, Django, Linux, Hadoop Sandbox, Apache Spark, Matlab, JS Bin EDUCATION

Master of Computer Science and Engineer (MajorGPA: 3.40) 0 8/2013 - 05/2016 University of Nevada, Reno, Nevada

Master of Applied Mathematics (MajorGPA: 3.72) 08/2012 - 05/2014 University of Nevada, Reno, Nevada

WORK EXPERIENCE

Web Developer, Arecy LLC, New York, NY 09/2016 - Present Personalized Restaurant Recommendation Web Service GitHub: h ttps://github.com/universehuayuzhou/restaurant

● Developed an interactive web page (H TML5/JavaScript) for users to search restaurants, update preference and view recommended restaurants. Implemented a web service using (J ava servlet, RESTful API) to fetch restaurant data from Yelp API.

● Utilized M ySQL/MongoDB to store user preference and restaurant information. Designed and developed a filtering and sorting algorithm and matched similar restaurants.

● Developed a collaborative Android app for users to search restaurants, update preference and view recommended restaurants (ListView, MapView).

● Tested the web service and app with unit tests ( JUnit ).

● Deployed server side to A mazon EC2, t ested by Apache JMeter .

Environment: Eclipse, Apache Tomcat, MAMP, Android Studio, MongoDB, Amazon EC2, JS Bin. PROJECTS

Machine Learning Application - Customer Churn Prediction in Telecommunications Industry

● Applied supervised learning models (logistic regression, random forest, etc.) to identify customers who are likely to stop using service in the future.

● Used different feature selection methods to analyze top factors that influence user intention in telecommunication companies.

● Implemented this machine learning pipeline using A pache Spark ML-lib and test it on Hadoop ecosystem.

Environment: H adoop Sandbox, Apache Spark.

Natural Language Processing – Movie Review Documents Analysis and Topic Modeling

● Applied Natural Language Processing methods (TF-IDF, N-grams, etc.) to cluster unlabeled documents into different groups and visualize results.

● Identified latent structures from documents using different clustering models (K-means, Latent Dirichlet Allocation).

● Visualized model training results by dimensionality reduction using Principal Component Analysis.

Contact this candidate