https://github.com/davgiles/ut-austin-data-science-program

This repository contains my projects from the Data Science & Business Analytics Post-Graduate Program through UT Austin.
https://github.com/davgiles/ut-austin-data-science-program

eda matplotlib numpy pandas python scikit-learn scipy seaborn visualization xgboost

Last synced: 19 days ago
JSON representation

This repository contains my projects from the Data Science & Business Analytics Post-Graduate Program through UT Austin.

• Host: GitHub
• URL: https://github.com/davgiles/ut-austin-data-science-program
• Owner: davgiles
• Created: 2024-08-02T04:42:28.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-08-27T14:46:32.000Z (4 months ago)
• Last Synced: 2024-11-10T04:06:42.370Z (2 months ago)
• Topics: eda, matplotlib, numpy, pandas, python, scikit-learn, scipy, seaborn, visualization, xgboost
• Language: Jupyter Notebook
• Homepage:
• Size: 9.31 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# UT Austin Data Science Program

![McCombs School of Business](https://www.mccombs.utexas.edu/media/mccombs-website/site-assets/images/utilityNav-logo.svg)

---

Throughout this program, I've learned data visualization and exploratory data analysis techniques through the Seaborn and Matplotlib libraries, statistical functions with the SciPy.stats library, building supervised learning models, handling imbalanced data, and hyperparameter tuning with the Imbalanced-Learn, Scikit-Learn, and XGBoost libraries.

Profile: https://eportfolio.mygreatlearning.com/david-giles

---

## Projects:

### [Project 1 - Data Visualization](Project1-EDA.ipynb)

- This notebook analyzes food order data to uncover trends in restaurant demand. By examining order frequency, customer ratings, and delivery metrics, the analysis aims to provide actionable insights to enhance customer experience and optimize restaurant partnerships.

### [Project 2 - Business Statistics](Project2-BusinessStatistics.ipynb)

- This notebook evaluates the impact of a redesigned landing page on user engagement and subscription rates for a news website. By comparing user interaction and conversion metrics between the existing and new landing pages, the analysis aims to determine if the new design attracts more subscribers and improves user experience, while also exploring variations in these metrics across different language preferences.

### [Project 3 - Supervised Learning (Regression)](Project3-LinearRegression.ipynb)

- This notebook develops a linear regression model to predict the pricing of used phones and tablets for a used phone store, a startup in the refurbished device market. By analyzing key factors that influence device prices, the model aims to provide a dynamic pricing strategy to optimize revenue and competitiveness in the growing used device market.

### [Project 4 - Supervised Learning (Classification)](Project4-Classification.ipynb)

- This notebook focuses on building a machine learning model to predict booking cancellations for a hotel chain. By analyzing factors influencing cancellations and developing a predictive model, the project aims to help the hotel chain reduce revenue loss, optimize resource management, and create effective cancellation policies.

### [Project 5 - Ensemble Techniques](Project5-EnsembleTechniques.ipynb)

- This notebook develops a machine learning classification model to assist the processing of visa applications. By analyzing historical data, the model aims to predict the likelihood of visa approval and recommend whether applicants should be certified or denied, streamlining the review process and improving efficiency.

### [Project 6 - Model Tuning](Project6-ModelTuning.ipynb)

- This notebook develops and tunes various classification models to predict wind turbine generator failures for an energy company. Using sensor data, the models aim to identify potential failures early, allowing for preemptive repairs and reducing overall maintenance costs by minimizing the impact of undetected failures and optimizing inspection and repair processes.

### [Project 7 - Clustering](Project7-Clustering.ipynb)

- This project aims to assist a financial consultancy firm in developing personalized investment strategies by performing cluster analysis on stock market data. The dataset includes stock prices and various financial indicators for companies listed on the New York Stock Exchange. The analysis focuses on grouping stocks based on similar characteristics and minimal correlation to create diversified portfolios. By identifying clusters of stocks with similar attributes, the project helps investors better understand different market segments, manage risk, and optimize their investment portfolios for steady returns, even in volatile market conditions.