Training Machine Learning Models On 311, 511, and 911 City Data

We have been working hard to understand the core stack of data services that make our cities work, or not work, depending on where you live. While we have labeled this research “smart cities”, we are starting with the basics of open data required for city operations. This is the current data sets available via existing services, which may or may not exist in a machine readable format, via an API, depending on the city you live in. There is a huge amount of city data already available at the municipal level, but here is where we have started as of January.

311 – Non-emergency Events

What is 311 Data and Why is it Important?
– Real Time Streaming 311 Incidents In Chicago
– A Demo

511 – Traffic, Travel & Transit

Adding 511 Data To Our Existing Transit Data Research
– Getting Your 511 Traffic Incidents in the San Francisco Bay Area as a Real Time Streaming API
– A Demo

911 – Emergency Events

Making 911 Data Real Time
– Streaming 911 Emergency Data For Baltimore, MD
– A Demo

We’ve targeted these three areas because they make a difference in our lives at the local level, and have huge potential when it comes to making available via web APIs, and in real time using Server-Sent Events (SSE). Now that we have these three critical aspects of municipal operations profiled, we are going to work to profile as many cities as we can. As we got started, we have been looking for actual 311, 511, or 911 APIs. After that we are looking for RSS feeds, and other ways we can publish APIs by scraping or other means–whatever it takes to get up some simple JSON APIs that we can turn into real time feeds using Streamdata.io.

While the primary objective with this work is to increase the number of city data feeds available, and turn into city data streams, a secondary motivation is to be able to take these real time streams and begin train machine learning models using the data. We are having conversations with our partners to better understand how Streamdata.io can deliver real time city data to them, so they can keep their machine learning (ML) models up to date in real time. As we progress in our work, we will publish here to the blog, and we’ll add any new APIs and data streams available to the Streamdata.io API Gallery, so that you can put to work in your own projects.

**Original source: streamdata.io blog

LEAVE A REPLY

Please enter your comment!
Please enter your name here