Diane's repositories
UR-Excel-Analysis-Challenge
Kickstart My Chart : Uncovering trends in Kickstarter projects, in an attempt to discover some tips & tricks for a successful campaign. Used Microsoft Excel for analysis & visualization.
UR-Big-Data-Challenge
Amazon Vine Review Analysis: Used Apache Spark & PySpark to ETL big data from Amazon’s S3 buckets into a new AWS RDS database. Used SQL to connect to RDS database, manipulate data, & analyze whether reviews from Amazon’s Vine program are trustworthy.
UR-Deep-Learning-Challenge
Charity Funding Predictor Neural Network: Creating a machine-learning algorithm to predict whether recipients of funding from the charity will be successful. Pre-processed data including one-hot encoding and scaling. Compiled, trained & evaluated a neural network model using Tensorflow. Finally, attempted model optimization efforts using kerastuner.
UR-Leaflet-Challenge
Visualizing Earthquakes : This project uses leaflet to visualize recent earthquake data from the USGS in terms of magnitude and depth. Using d3 to download the current json data, a live interactive version of this map can be hosted on a website.
UR-Matplotlib-Scipy-Stats-Challenge
Pymaceuticals Plotting : Performed various analysis and visualization of data relating to [mock] cancer drugs in consideration for treatment. This was completed using Python, Pandas, and Matplotlib.
UR-Project1-COVID-and-Stock-Market
Exploring Effects of Pandemic on Stock Market : Specifically, looking at relationships between caseloads and stock value, trading volume, and price amplitude, among others.
UR-SQLAlchemy-Challenge
Used Python & SQLAlchemy to perform analysis on climate data from NOAA weather stations in Hawaii. Then created a Flask API to return JSON data from this analysis.
UR-Tableau-Challenge
Citi Bikes Analysis: Performed ETL on multitude of datasets provided by Citi Bikes for years 2019 & 2021 using Python/Pandas. Final loaded dataset after sampling contained nearly 10 million records. Performed additional manipulations & analysis using Tableau to investigate effects of the COVID-19 pandemic on rider behavior. Created Tableau dashboards with brand-oriented visualizations and a cohesive Tableau story presentation
UR-Web-Scraping-Challenge
Mission to Mars: Built Flask application that scrapes various websites for data related to the Mission to Mars using BeautifulSoup & Pandas, stores the data in a MongoDB database using Pymongo, and then renders the data from MongoDB into a new web page created using HTML & Bootstrap.
UR-Pandas-Challenge
Heroes of Pymoli : Using Python & Pandas, analyzed data for a gaming company, and generated a report focused on turning the in-game purchasing data into meaningful insights.
UR-Project2-Police-Shootings-ETL
Performed ETL on data collected from various sources (csvs downloaded from various sources and the US Census API) to create a single database, in order to explore questions surrounding the "Defund the Police" movement. Specifically looking at data regarding police-involved shootings and state budgets.
UR-Project3-Vacation-Explorer
Creating a comprehensive dashboard to visualize various travel concerns in one place to ease the process of planning a trip. Primary feature of this dashboard is a leaflet map visualizing COVID-19 case rates, climate info, and tourist attractions for the 50 of the most popular destinations within the US.
UR-Project4-Predicting-Housing-Market
Created and evaluated a linear regression algorithm to attempt to predict the average cost of the housing market (ZHVI from Zillow) in given metropolitan statistical areas. Compiled model using features including the federal funds rate and MSA demographics and market features such as unoccupied housing and net migration.
UR-Python-API-Challenge
Visualizing Weather : Using Python to download json data relating to weather in cities around the world into a pandas dataframe. The data was explored and visualized in various manners including regression. Finally, the weather data was plotted on a gmaps heat map and narrowed down to select cities meeting the author's ideal weather criteria for which to include additional destination information from the Google Places API.
UR-Python-Challenge
PyBank & PyPoll : Created Python scripts to perform a financial analysis and an election results analyis.
UR-SQL-Challenge
Employee Database Engineering & Analysis : Created a data model and engineered a SQL database for employee information. Then performed SQL analysis to complete various employee reports.
UR-Supervised-Machine-Learning-Challenge
Predicting Credit Risk : Created a machine-learning model that attempts to predict where a loan becomes high risk or not. Both a logistic-regression model & random forest classifier were compiled and evaluated.
UR-Unsupervised-Machine-Learning-Challenge
Cryptocurrency Clusters : Can cryptocurrencies be grouped together under a classification system to be used for an investment portfolio? Used Python to pre-process raw data, including one-hot encoding and scaling. Attempted dimensionality reduction using PCA & t-SNE. Then performed a k-means clustering analysis to attempt to identify appropriate clusters.
UR-VBA-Scripting-Analysis-Challenge
The VBA of Wall Street : Created VBA script to loop through all the data in multiple excel worksheets which outputs an analysis of aggregate data for each stock ticker