Online Supermarket Management System Boulder, United States Course Project Mar.2020-May.2020 • Built an online supermarket system with functions of searching, Adding and placing order for customers and Adding, removing, checking items for staff using Java as backend tool • Using mysql as database and html, javascript as Frontend tools. Connecting Java and database by JDBC and using Tomcat as http web server
Movie Recommendation System in Apache Spark Boulder, United States Individual Work Nov.2019-Feb.2020 • Built a recommender engine using Apache Spark and Elasticsearch and trained the model using matrix factorization method as well as alternating least square techniques • Conducted model hyper-parameters tuning with Spark ML cross-evaluation toolbox and monitored data preprocessing performance via Spark UI on AWS • Used Elasticsearch to persist the trained model, query, score, rank and display results.
Data Mining: Customer Churn Prediction in Telecommunication Industry Boulder, United States Individual Work Oct. 2018-Dec.2018 • Preprocessed dataset of by data cleaning, categorical feature transformation and standardization. • Trained supervised machine learning models including Logistic Regression, Random Forest and K-Nearest Neighbors • Evaluated model performance by k-fold cross-validation technique and analyzed feature importance to identify top factors that influenced whether the customer would stop using the service