Hey there, I'm Emily - a systems engineer turned aspiring data scientist. Welcome to my data science portfolio. Through my portfolio, I aim to showcase my ability to use machine learning tools to extract insights from data and present them through impactful visualization.
“Practice isn't the thing you do once you're good. It's the thing you do that makes you good."
― Malcolm Gladwell, Outliers: The Story of Success”
Summary
The project aimed to assess customer experience by analyzing Google Maps reviews for Starbucks locations in the western district of Hong Kong Island. To overcome the limitations of the Google Maps API, Selenium and BeautifulSoup were used to extract more than 5 reviews per location. The review data was analyzed using the GPT-3.5-turbo model through OpenAI API.
The insights drawn from the project suggest that analyzing review content provides a more nuanced understanding of customer sentiment and experiences, and highlights the need for location-specific analysis and improvements to address issues contributing to negative customer experience.
Read More →
GitHub (View Code) →
Summary
An LSTM model was built and trained on historical AAPL stock prices from Yahoo Finance to predict the stock prices for AAPL in 2023.
Two LSTM models were compared after fine-tuning the hyperparameters, and the results showed that a simple single-layer LSTM was better at modeling the actual stock price data and forecasting than the multi-layer LSTM model.
Overall, the LSTM model can be effective for predicting stock prices or other time-dependent data after careful consideration of model architecture and hyperparameters.
Read More →
GitHub (View Code) →
Summary
The project is focused on helping new Airbnb hosts in Sydney gather valuable insights and strategies to maximize guest bookings.
The goal of the visualization system is to empower hosts with the key factors that drive bookings and ensure that they can run a profitable and sustainable accommodations.
Read More →
Tableau (View Dashboard) →