Hi, I am

Weishan.

I am seeking job opportunities in the data world!

I am a results-oriented data scientist with a proven track record of leveraging Machine Learning techniques to build impactful products and drive data-driven analyses. Proficient in A/B Testing, Data Management, and Data Visualization. Skilled in Python (PyTorch, NumPy, Pandas, Statsmodels, Scikit-learn, Seaborn, Matplotlib, etc.), SQL, Tableau, Looker, PowerBI, R, Git, Poetry, Docker, Google Cloud Platform, AWS, Hadoop/PySpark, and Web Scraping.

In my recent role as a Data Scientist at Andela, I successfully developed and deployed a recommendation system using tree-based models and NLP. This system effectively matches talents with suitable job opportunities. Additionally, I led experimentation efforts, conducting A/B testing on user flow and machine learning algorithms. Prior to Andela, I held positions as a quantitative research analyst and senior business analyst at two marketing agencies. In these roles, I utilized statistical analysis on structured and unstructured data to help client companies gain valuable insights into customer behavior and measure marketing efficiency.

I am continuously committed to enhance my data science skill set. In my spare time, I avidly read articles and books, and engage in projects that further develop my expertise. Currently, I am focused on expanding my knowledge in NLP and MLOps. Check out my Projects and Blogs for more cool stuff!

Projects
Resume
      

Projects

Stock Price Prediction with LSTM
LSTM Time-Series PyTorch
Stock Price Prediction with LSTM
Applied LSTM on time-series data to predict future stock price
Demo
Build Hotel Database Using Web Scrapping
Web Scrapping ETL MongoDB
Build Hotel Database Using Web Scrapping
Leverage Selenium and BeautifulSoup to scrap hotel information from booking.com and convert the data into MongoDB collection to save them in a JSON file
Demo
New Feature Performance Evaluation using DiD and CLV
DiD CLV R
New Feature Performance Evaluation using DiD and CLV
Use diff-in-diff and logistic regression to evaluate the effectiveness of a new feature - online gaiming community - by measuring the change in user revenue and retention
Demo
New TV Design Using Conjoint Analysis
Conjoint Analysis Pricing Analysis
New TV Design Using Conjoint Analysis
Deploy conjoint analysis to generate a TV Design that will lead to the optimal market share
Demo
Superstore Sales Performance Dashboard
Data Visualization Tableau Dashboard
Superstore Sales Performance Dashboard
Construct an interactive dashboard in Tableau to track sales performance for a supermarket
Demo
Use Bootstrap to Obtain Confidence Levels for Willingness to Pay
Bootstrap
Use Bootstrap to Obtain Confidence Levels for Willingness to Pay
Apply Data Bootstrap and Residual Bootstrap to obtain the 95% confidence levels and median for willingness to pay
Demo
Use PCA and PCR to Visualize Brand Competition in Brand Map
PCA PCR Brand Map
Use PCA and PCR to Visualize Brand Competition in Brand Map
Build brand maps for car brands using principal component analysis and recommend strategies to focal brand's mangers to improve their product design
Demo
Fraud Detection Using Supervised Machine Learning
Logistic Regression Random Forest XGBoost Python
Fraud Detection Using Supervised Machine Learning
Build ETL and apply logistic regression, random forest, and XGBoost to detect fraudulent activities
Demo