PROJECTS
J.D. Power: Predicting Sales in 2021
This project is contributed via UC Irvine Capstone Course during my Master's in Business Analytics program. I worked in a team of 5 with J.D Power to build models for predicting the auto industry sales in 2021 under the affecting of COVID-19.
The goal of this project is to provide the ideal BlackBox for models to predict vehicles sales via historical data and provide comprehensive recommendations.
-
Model: Time Series Analysis, Regression, ARIMA with X-REG
-
Propositions: We proposed to develop new segments for Electric and Hybrid based on the fastest-growing
-
Skills Used: R, Data Cleaning, Tableau, A/B Testing
Wildtrack: Recommendation - UI/UX&Data
This project is contributed via Seed Consulting Group to provide the suggestion of UI/UX Design, data management, data presenting interface, and stabilize funding sources for Wildtrack.org, a non-profit organization, collecting wild animal trace data to protect wild species.
I worked along with four members, including the project manager. I built the demo website through the WIX.com website builder to design the website interface, and we provided the following recommendations for stable funding.
Propositions of funding sources:
-
Partnerships
-
App and site monetization - donations and subscription
-
Advertising
-
Eco-tourism
-
Online learning and certifications
-
Student programming
-
Membership
-
Crowdfunding
-
Shareable/virtual fundraising campaigns
-
Selling of data
-
Government
Skills Used: UI/UX, Project management, Strategy, Wixsite.com
Flight Delay Prediction
This project is associated with the course of predicting analytics I have taken during my Master's in Business Analytics program.
The goal of this project is to build models to predict whether flights will be delayed via US Historical Flight Delay and Weather Data and provide comprehensive recommendations and key takeaways.
We chose the data in December 2019 to predicting delay since December has a higher demand for flights during the Christmas holiday, New Year, and Winter break.
-
Key Takeaways: Avoid carriers, which has higher delay rate, such as Jetblue, Southwest Airlines, Alaska Airlines
-
Model: Logistic regression, Decision Tree, Random Forest
-
Skills Used: R Coding, Data Cleaning, Tableau
Online Advertising Analytics
For Paul Merage School of Business
-
Key Takeaways and Propositions:
-
Higher cost of conversions than industry benchmarks
-
Have a tailored MBA Program and user-friendly website to attract more international students for MBAs programs, and more local students for 1-year specialty programs.
-
Diversify channels or change strategy to focus on conversion in google ads
-
-
Model: Chi-sq Test, Logistic Regression
-
Skills Used: R Coding, Data Cleaning, Tableau
Describe your image
Describe your image
Describe your image
Describe your image
This project is analyzing the dataset and evaluates the google ads effectiveness of Paul Merage School of Business at UC Irvine.
The goal is to provide practical suggestions to improve the digital
advertising strategies to attract more applicants through this project.
We mainly focus on application data for MBA and Speciality Master's programs, as well as various social media campaigns to promote these programs.
Flight Database Management
This project is associated with the course of business data management I have taken during my Master's in Business Analytics program to analyze flight delay data in December 2015.
Most people took long vacations during Christmas and New Year, we want to see how holiday influences flights. Therefore, the goal of this project is to know more about the flight schedules from different airlines.
Key Takeaways and Propositions:
-
Only 1.7% of flights in December are canceled, and 70% of cancellations are due to weather.
-
Alaska Airlines and Hawaii Airlines have the shortest average delay time, while the average delay time of United Airlines and Frontier Airlines is over 1 hour.
-
Choose the day around the end of a week (Thursday, Friday, Saturday) to lower cancel probability.
-
For further scope, we hope to analyze how to help people who love traveling during a holiday to choose the best routes with the best date and time and the best airline to avoid flight cancel and delay.
Skills Used: SQL, MYSQL, Google Cloud(Big Query)
Steam Gamer's Behavior Under COVID-19
The goal of this project is to prove that the stay-at-home order during the pandemic has an impact on the way people interact with each other, and the online gaming industry.
Data Processing:
-
Obtained data on the number of players recorded at the beginning of each day from Steam DB as game activity data.
-
Extracted the sample of 8,000 games from Steam SPY API.
Key Takeaways:
-
Increase in engagement and revenue as a result of COVID-19.
-
Games with the keywords of Action, Adventure, Multiplayer, Free to Play, and Co-op have more activity users.
-
A change in social behavior that people are turning to interact by virtual because of COVID-19.
Skills Used: Python, Juyper Notebook, Steam SPY API
Describe your image
Describe your image
Describe your image
Describe your image
Online Shopper's Intention
This project is associated with the course of machine learning I have taken during my Master's in Business Analytics program.
Online retail consumer behavior has become crucial in analyzing factors that influence online buying behavior to find ways to increase company revenue.
The goal of this project is to find the factors that affect conversion and customer intention and further effectively increase company revenue.
Key Takeaways and Propositions:
-
The most important attribute is "Page Value."
-
Based on the clustering results, users who spend more time browsing the website have the lowest bounce rate.
-
Optimization of Landing pages and creating a user-friendly interface.
-
Personalized targeted emails and loyalty programs for returning customers to increase sales and revenue.
Skills Used: Machine Learning (Clustering, Decision Trees, Random Forest), Weka
Data360 Hackathon: Health Alert
This Hackathon is associated with Data360 I participated in during my Master's in Business Analytics program.
The goal of this Hackathon is to address any COVID-19 related issues via dataset, including posts with unstructured text from Facebook and Twitter.
I worked in a team of two. The goal of our team is to provide immediate help to those in need, especially in areas affected by the epidemic. Thus, the purpose of this project is early real-time detection of COVID-19’s vaccine side effects in various locations.
Skills Used: Python, Tableau
Peloton Interactive Inc.
Marketing Investment Memo
This project is associated with the Foundations of Marketing course I have taken during my Master's in Business Analytics program.
This assignment is to write an investment memo for a company. I chose Peloton Interactive Inc, which has become a popular at-home brand during quarantine. Peloton provides both hardware and software services, which combines advanced design and modern technology that bring revolutionary changes to the fitness industry. It helped me to learn how Peloton work and how the company earns and technologically.
