Transforming the Digital Landscape: A Fusion of Business Insight and AI Expertise

Intro

Welcome to my professional realm, where I blend advanced AI and analytics with strategic business acumen. As a Management Consultant at Kearney Analytics, my expertise lies in converting complex challenges into analytical solutions. My focus on Generative AI and NLP have notably advanced business applications in this cutting-edge area, reflecting my commitment to both technical excellence and commercial viability. I am passionate about Large Language Models, Optimization and AI.

Beyond the boardroom, I enjoy traveling, painting, playing board games and badminton, and cooking. So far, I've visited Spain, the United States, Japan, Mexico, South Korea, China, Singapore, Indonesia (Bali), the United Kingdom, the Czech Republic, Austria, United Arab Emirates (Dubai), and Thailand!

Fun fact: I am a proud Google Maps Level 7 Local Guide. My shared photos amassing 9,066,035 views.

Publication

Test Certification

Personal Development Certification

Award

🌈 I am Paul

Data Enthusiast, Artist, and Boardgame Lover

🙌 My Skills

 
Python SQL Excel VBA PowerBI HTML SAS

🎓 My Education

 
Purdue University Krannert School of Management

Master of Science in Business Analytics and Information Management

May 2022

  • Academic Honors | Merit Scholarship Recipient
  • Technical Coursework: Computational Complexity, Analyzing Unstructured Data, R for Analytics, Data Mining, Advanced Business Analytics, Business Analytics, Management of Organizational Data, Prescriptive Analytics, Python, Speadsheet Modeling
  • Business Coursework: IT Project Management, Customer Analytics, IT Innovation, Change Management, Communication and Persuasion, Operation Management
National Cheng Kung University

Bachelor of Management Science, Industrial and Information Management

Jun 2020

  • Relevent Coursework: Operation Research, Optimization, Invertory Management, Quality Management, Marketing Management
  • Capstone Project – “Assembly line balancing problem - The case study of a shoe manufacturing company”

🤝🏼 My Projects

 

Apple Watch Analysis
Bulding a dashboard webpage via R and Shiny using Apple Watch data. This dashboard shows my excersie and sleep quality across eight months.
Here is the webpage.
R Shiny, Data Visualization

SIL Machine Translation Seed Optimization
SIL is an NGO dedicated to preserving languages. Building translation models require a huge amount of data. However, in low resource language, SIL needs to hire humans to translate the data. We create an effective method to reduce monetary cost and training time for translation models by scripting an algorithm to optimize training datasets and provide a clear strategy for selecting semantic domains (translation content) via Python. That is, we can use a smaller amount of datasets to achieve the same performance.
Python, Neural machine translation

Integrating prediction models into optimization models
Many small businesses suffer from how to leverage data to improve performance. However, hiring data specialists is challenging for small businesses. We provide a user-friendly framework that integrates the prediction model into the optimization model. Small businesses can use this autoML framework to get heuristic solutions.
Python, AutoML, Optimization

SAS Optimization Challenge
Aggregated data and used Python library autots to forecast the water usage. Then, constructed 3 models via SAS to determine which contract the corporation should adopt. Won the 1st place in this challenge.
Here is the demonstrated video.
Python, SAS, Optimization

Craigslist - Video Gaming Classification
Scrapped, tokenized, and lemmatized the Craigslist video game data, then built 6 models with 88.45% accuracy to increase information gain by 34% which improved the retention rate and customer experience in Craigslist.
Python, NLP

Crossroads Classic Datathon - Phishing Email
Wrangled and tokenized publicly available data as well as an internal dataset to build a model with 98.53% accuracy that can identify phishing emails. This model wins 3rd place in Kaggle Leaderboard.
Here is the demonstrated video.
Deep Learning, Python

Taiwan Sanoh Database Project
Utilized Excel VBA to preprocess large datasets and designed the ER diagram for building the database. Normalized the datasets and used MySQL to dig business value from data.
Excel VBA, SQL, Database

Heart Disease Classification
Built a patient friendly R shiny app which is a website that allows users to input patient body data and predict the potential heart failure level.
Here is the demonstrated video.
R

Bike Sharing Analysis
Utilizing regression to predict numbers of users, using stratified sampling and XGBoost to predict user type with 84.04% accuracy, and manipulating network and graph algorithms to offer useful insights.
Python

Kaggle - Credit Card Prediction Project
Created and tuned an ensemble model that combined neural network, adaptive lasso, and random forest through SAS EM to predict the likeliness of credit default and won 10th place in the Kaggle in-class competition.
SAS

Coursera Analysis
Cleaned, integrated, manipulated and visualized Coursera datasets to find critical elements and offer business recommendations to enhance course enrolment using statistics models like one-way ANOVA, OLS, and regression.
Python

Bellabeat Analysis - Case Study in Google Data Analytics Certificate
Gathered, cleaned, and analyzed FitBit dataset for offering data insights.
Python

Assembly Line Balancing Problem- The case study of a shoe manufacturing company
Created a model that can optimize the assembly line layout in a matter of seconds. Cleaned the data and made it readable with Excel VBA and coded in Python and used Gurobi to optimize the result of this project.
Python, Gurobi, Excel VBA

Boosting Recycle Rate in China
Leveraged design thinking methods and cooperated with students from various countries to conduct independent studies.
Design thinking