Data Science & Software Engineering Portfolio

Jeremy Seibert's portfolio of of Data science, Econometric, and Software Engineering projects completed for academic and self-learning purposes. Each of the notebooks will take you to an Jupyter NBViewer, and the proceeding link will take you to the Github Repository directly.

Contents

• Software Engineering

  • Jeremy Seibert website Check out my personal website for more information about myself, such as my resume, background and more about my projects. I built the site using React, Framer-Motion, and Materialize-CSS!

  • Loop: (In-Progress) Loop is a platform for local storytellers to engage with their communities through quick stories. We are building the back-end infrastructure on top of GCP and Firebase, and front-end is being built using the React. This site is currently not live, but our launch date is targeted for December 2020.

  • Sorting Algorithm Visualizer This is a simple a sorting algorithm visualizer. Currently it has the quick sort and bubble sort algorithms implemented. Selection sort, insertion sort, heap sort, and merge sort algorithms are in the build pipeline.

  • QualtricsAPI : Qualtrics is an awesome company that builds software which gives users the ability to collect online data through online surveys. This python package, exists as a wrapper on top of the Qualtrics API. This package's primary goal is to be a super convenient way for python users to ingest, or upload their data from Qualtrics to their development environment, and vice versa. I do want to make point to say this package is not affiliated with Qualtrics, and is an open-source project.

  • Spaceman3D: Spaceman3D is a python package that accomplishes several unique tasks for Python users working within the field of Astrodynamics. The package gives users the ability to parse satellite Two-Line Element (TLE) Data into its Ballistic, Keplerian, and Identifying elements. Beyond, this Spaceman3D uses a Matplotlib 3D plotting toolkit to plot the trajectory and orbit of the satellites relative to the earth Pip install spaceman3D. PyPi

  • Federal: This is a simple package built on top of pandas datareader to make it easier to pull in Economic Data from the Federal Reserve in St. Louis's Federal Reserve Economic Data (FRED) tool. Pip install Federal. PyPi.

  Tools & Languages: JavaScript, TypeScript, Python

• Economics & Mathematics

  • Modeling Production Efficiency: Using the Edgeworth Box: This is a demonstration of the how to model the Pareto Effecient production Capital and Labor outcomes using the Edgeworth box and MicroEconomic Theory. Github

  • Zipf's Law and US Metro Population Growth: This is a demonstration of the Zipfs Law (i.e. the Rank-Size Rule) using the 2017 Population Estimates supplied from the U.S. Census Bureau. Github

  • The Mathematics of the Powerball Lottery: This is the notebook associated with the December, 27th 2018 Urban Scientist Blog Post. Check out the Urban Scientist www.urbanscientist.cool.

  Tools: scikit-learn, Pandas, Seaborn, Matplotlib

• GeoSpatial Analytics

  • Geospatial and Exploratory Analysis of Starbucks Locations: Exploratory Analysis of the Starbuck's Store Directory dataset, with a build of a mapping tool that plots each store within a user-defined city our country. The mapping tool that I use is folium, a python wrapper on Leaflet.js. Github

  • Geospatial Presentation: This is the notebook behind my May 15th Greater Evansville Data Science Presentation. It covers several unique geospatial packages in the python ecosystem. Github

    Tools: folium, geoPandas, geopy

• Machine Learning & Data Engineering

  • Working with the Spotify API: This repository contains a notebook that explains how to use SpotiPy to gather data from the Spotify API. Github

  • Spotify Song Prediction: This notebook is the exploration between two Spotify playlists, one filled with songs that are liked, and the other filled with songs that are disliked. These playlists are populated with songs built around my song preferences. Our goal here will be to use several machine learning algorithms, to predict whether songs will be liked, and which songs will be disliked. Github

  Tools: sklearn, Keras, Pandas, Networkx, etc.

• Web Crawling & Scraping

  • Glassdoor Crawler: This repository contains a python package that aggregates specific job listings from glassdoor.com, ranks them according to specified user preferences, and then identifies the keywords/phrases used in the job description.

  Tools: Requests, Selenium, Pandas, NLTK, etc.

If you liked what you saw, want to have a chat with me about the portfolio, work opportunities, or collaboration, shoot an email at jaseibert5@gmail.com.