Predicting NFL Playoffs
Using positional salary spending and previous year stats to predict NFL playoff teams
Introduction Professional sports tend to have a lot of inertia. The best teams in one season tend to be contenders the next season as well. Although players get shuffled around through trades, free agency, drafts, and waivers, the full cast of players on each team doesn’t change that much from year to year. The NFL is a prime example of... [Read More]
Posted Jun 6, 2018
Building a Simple Recommender System in PySpark
Using PySpark to build a simple recommender system based on the MovieLens 100k reviews dataset
In this post, I will walk through a simple recommender system with PySpark from scratch using the MovieLens 100k ratings dataset. I take a novel, weighted approach based on number of similar reviews and cosine similarity. Part 0 - Setting up the Environment To start, setting up the environment to accommodate python, spark, and hadoop can be arduous. I have... [Read More]
Posted Mar 26, 2018
Clustering Neighborhood Change in Austin
Using census data and unsupervised learning to investigate socioeconomic change, price inequality, and gentrification
Part 1 - Project Proposal Part 2 - Data Wrangling Part 3 - Exploratory Data Analysis Part 4 - Choosing a Clustering Algorithm Part 5 - Clustering Analysis Part 6 - Closing Remarks Part 1 - Project Proposal Top Problem: Austin is my home, I love it here, and I plan on being here for as long as possible. The... [Read More]
Posted Mar 15, 2018
Model-Level Hydrodynamic Particle Tracking
Supporting oyster reef restoration projects in the Trinity-San Jacinto and Mission-Aransas Estuaries
All code for this project is located in the project repository HERE Introduction To support oyster reef restoration efforts and other potential projects (like oil spill response), I developed model-level particle tracking functionality for TxBLEND, a two-dimensional hydrodynamic and salinity transport model used by the Texas Water Development Board (TWDB) to simulate water circulation and salinity patterns in Texas estuaries.... [Read More]
Posted Dec 15, 2017
Building a Github Blog with Jekyll and Ruby
In this three-part tutorial I’ll walk you through how to build a Github blog from scratch with Jekyll and Ruby. While most of this is pretty straight forward, there were a few things I encountered that were a bit tricky to understand and/or implement. Note: I’m setting this up on OS X so a few of the commands are system... [Read More]
Posted Oct 1, 2017
subscribe via RSS