Pinned

1. Walmart-Weekly-Sales-Forecast Walmart-Weekly-Sales-Forecast Public

   We created four models, such as SARIMAX, Prophet, and Ensemble Model(Prophet+XGBoost), to forecast Walmart 45 stores' weekly sales.

   Jupyter Notebook 1

2. Customer_Churn_Prediction_Using_PySpark Customer_Churn_Prediction_Using_PySpark Public

   This project aims to leverage big data technologies to help OTT platforms predict churn at a customer level in real-time. We use Amazon S3 as the data storage, Databricks for data processing, spark…

   Jupyter Notebook 3

3. Causal-Inference_ROI-measurement-on-sponsored-Ads Causal-Inference_ROI-measurement-on-sponsored-Ads Public

   We performed the Difference-In-Difference analysis to check whether there is causal effect between sponsored search ads and organic clicks on Google search platform.

   HTML

4. Movie-Recommendation-System Movie-Recommendation-System Public

   We create movie recommendation system through demographic filtering, content-based filtering, collaborative filtering, and hybrid engine.

   Python

5. Spam-Email-Detection Spam-Email-Detection Public

   I built several classification models to detect whether an email is spam or not. I built models and found the best performance using Nested CV and GridSearch under two scenarios: Accuracy-Based & C…

   Jupyter Notebook 2

6. Consumer-Spending-Prediction Consumer-Spending-Prediction Public

   I built several predictive models to predict consumer spending, and conducted GridSearch to find the best hyperparameter for every models. Models I used including: Ensemble Stacking/ Deep Neural Ne…

   Jupyter Notebook