Used JavaScript and React Native to develop a mobile application that provides resources and connects them with mentors accordingly to a user-entered text describing the challenges they face
Aimed to analyze user-entered text using C++
Won second place
Personal Website, Personal project
May 2023
Developed a personal website using JavaScript and HTML/CSS: www.crystalzhao.com
Utilized Figma and created interactive prototypes for optimal user experience
Deployed using AWS to ensure accessibility to the website without a development environment and to optimize website performance
Parent Plan It, cmd-f 2023 Hackathon
Mar 2023
Used React Native and Expo to build a user-friendly front end of a mobile application to help single parents budget towards their financial goals with the right resources and goal-saving tracker
Utilized JavaScript to build an intuitive navigation system, personalized resources center, and user authentication features, as well as a user-friendly interface for mobile application
Won Best Figma Prototype and UBC CS Project Hub Award
Financial Recorder, Academic
Feb 2022- Mar 2022
Built a Java application that allows users to add, remove, view, modify, and delete their expenditures and savings, with a display of their current balances
Implemented data persistence using JSON to allow users to save and load from the application
Developed both console application and GUI using Java Swing Library and JFrame as user interfaces; achieved 100% test code coverage using the JUnit
Crime Rate and Economic Inequality Analysis, Academic
Mar 2023- Apr 2023
Used R to bootstrap the dataset, and constructed a confidence interval and hypothesis test to analyze the correlation between economic inequality and the crime rate in Vancouver between the years 2012 and 2016
Conducted analysis of crime rate patterns and proposed potential areas for future research
Heart Disease Prediction in Hungary, Academic
Oct 2022- Dec 2022
Used R to train the model to predict the prediction on whether a person is at risk of heart disease and achieved 79.8% accuracy
Built the classification model using the k-nearest neighbours method and choose k with the highest accuracy on validation data using cross-validation