Inspiration

We wanted to understand how the intersections of both big and small real drug trade happen and where they occur. We also wanted to be able to give a global perspective on how drug trade occurs in the U.S and other countries. Using analysis and data visualization tools, we wanted to observe correlations in areas severely effected by drug trafficking and high drug overdoses. We wanted to be able to provide a service that predicts where and when certain drug traffic moves and how it impacts the areas around it.

What it does

Analyzes raw data on drug overdoses in the U.S and using an intensity scale of points to identify how the overdoses in that area have increased/decreased over time. We then (attempt) to use the raw data and map-view to understand how some areas are more/less effected and correlate to increases or decreases of drug trafficking in their area.

How we built it

We used HTML/CSS to build a basic website and linked the project to a Torque.js modified SaaS mapping tool. We attempted to implement a machine learning algorithm to search and sort raw data for us and then parsed the data through Torque.js in order to display it to CartoDB.

Challenges we ran into

Finding (enough) open data on drug trafficking and implementing a web scraper tool and machine learning algorithm. We also tried multiple frameworks and many advanced visualization tools, but faced compatibility and complexity issues.

Accomplishments that we're proud of

We were able to provide dynamic visualizations of high drug overdoses around the country and show how they change over time. Amazing visualization.

What we learned

We learned how to incorporate CartoDB and raw data sets to create stunning data visualizations. We also learned how to use Torque.js for modifying said visualizations and manipulating data sets.

What's next for Spade Marlowe

We want to visually provide data on current drug traffic routes and create a browser-based modeling tool that the user can use to find out what areas are most effected (near them, perhaps) by direct/indirect en route drug trafficking, view the important correlations, and provide location data between specified cities or zip codes to see a more detail view of trafficking issues and get a model of how those issues will evolve over time in the specified area.

Built With

Share this project:

Updates