As the winner of the 2017 Seattle Global AI Hackathon, we created a fake news detector based on a Naive Bayes Classifier that could predict whether an article was fake with 85% accuracy. The model was trained on a set of 4,000 articles and was
Vehicle cruising (individuals looking for parking and for-hire vehicles operating without a passenger) is a major contributor to traffic congestion in downtown Seattle. Still, the magnitude and location of vehicle cruising is poorly understood.
To get a better understanding of where vehicles cruise, we used traffic sensor data to generate most-likely paths traversed. We segmentated paths by time and method of transportation. After segmentation, we introduced metadata to describe
the trip and used a semi-supervised machine learning approach to label the data. Ultimately, we created an interactive heat map of downtown Seattle that can be used to visualize the relative levels of cruising.
This research has the potential to help transportation agencies, technology companies, and car companies predict the availability of parking and more accurately direct travelers with online, mobile, and connected tools, thereby reducing congestion
impacts, emissions, and fuel costs. Check out source code and a
As a Seattle Angel Hack Challenge Winner, Basic Needs is an SMS-based application that provides access to routing, emergency, mental health, sanitation, education, location, weather, and feedback services to vulnerable populations without wifi
dependency. Basic Needs addresses immediate needs, connects vulnerable populations to their needs, and initiates a dialog between cities and vulnerable populations for social engagement and improvement.
Source code is available and showcases Python utilized in conjunction Amazon Lexa and AWS Lambda.
This visualization, second place winner, shows the disparity between access and need for food banks in Ohio. It was created by overlaying two different sources of data to see emerging trends. To create the drive time data, first, a list of census
tracts with center coordinates were produced. Then Python code was run to calculate driving time from that center coordinate to the nearest foodbank. This was achieved by utilizing Open Source Routing Machine. Next, we received SNAP data about
food insecurity from the Census. Our final step was overlaying these two data sources in QGIS to produce a meaningful visualization. See here for source code.
This presentation, winner of best visualizations, shows insight into the data provided for the 2017 Datafest competition. We analyzed Expedia data and provided recommendations for customizing pricing based on where you book.