GitHub Analysis - Data Science Project Aug. 2019​ to Sept. 2019​Analysis of GitHub statistics on repositories exploring what leads to a successful repoPython | Pandas | Numpy | SKlearn | NLTK |GraphQL- Scheduled data querying within limitations of GitHub GraphQL api- Feature engineered numerical time series data to run dynamic time warping- DTW results were then clustered to find repos that had similar trends - Ran sentiment analysis on text data to determine meaningful differences between the clustersMaster Your Money - Cross Functional Web App June 2019​Worked with a cross functional team to create a web app that predicts a score based on luxury vs necessity spending within earning bracket of userPython | Pandas | Numpy | SKlearn- Gathered data from bureau of labor statistics - Carefully looked over documentation to make sense of the data- Cleaned and encoded useful features, bucketed sets of data based on earning bracket - Ran random forest model to classify spending behavior Google Play Store - Data Storytelling Apr. 2019​Data exploration and data visualization of Google Play store data setPython | Pandas | Matplotlib | Google Co-lab- Imported data as Pandas dataframe- Explored the data to understand the data and find interesting talking points- Used descriptive statistics in exploring the data- Created data visualizations to facilitate the story that was being toldTanzanian Ministry of Water, Predict WaterPump May 2019​Accurately predict faulty water pumps using machine learning modelsPython | Pandas | Numpy | Sklearn | Matplotlib | Scipy | XGBoost- Cleaned data, encoded categorical features, feature engineered- Used random forest classifier and XGboost classifier and see which had higher test accuracy- Hyper-parameter tuned the better preforming model