Skip to content

Instantly share code, notes, and snippets.

@stuartlynn
Created July 14, 2016 14:46
Show Gist options
  • Select an option

  • Save stuartlynn/6b09f09f72cdd040a098f4fc832c14fa to your computer and use it in GitHub Desktop.

Select an option

Save stuartlynn/6b09f09f72cdd040a098f4fc832c14fa to your computer and use it in GitHub Desktop.
Thinking Spatially: How location intelligence can enhance your data and let you uncover new insights.

Talk Type

Learning Lab - 60 minute hands on skillshare or small group discussion (technical or other) Fireside chat - 45 minute talk with time for questions and answers

Session title

Thinking Spatially: How location intelligence can enhance your data and let you uncover new insights.

Describe what your session or activity will allow people to make, learn or do in 150 words or less

Location data has the fairly unique ability to be easily joined with a large number of other datasets at the same location. The fact that we have a latitude and longitude, or region or even just an address for a piece of data allows us to easily bring every other dataset that co-occupies that space into relation with our own data. This makes it incredibly powerful for developing deeper insights in to that data. In addition, most processes in our world have a spatial component, birds flock, diseases spread and your neighbors effect you. Thinking spatially about how the world, applying spatial statistics and prediction methods helps us interpret
our data better.

In this talk I will use a series of open source tools to show how you can easily join your spatial data with other datasets, use that to run spatial statistics models and machine learning based prediction.

Describe how you see that working in 150 words or less.

The workshop will kick off with a very brief discussion of what datasets are spatial and how that allows them to be easily related to datasets like the census. We will walk through an example of taking a dataset and joining it with many others and creating a explorable dashboard of the relationship between the datasets

Then we will spend a little time talking about basic spatial statistics and looking at an open source python package called PySal which will let people easily apply them.

We will work through an example of how to do this

Finally we will briefly touch on machine learning and how it applies to location and give some examples of projects where people are using it to good effect.

How will you accommodate varying numbers of participants in your session? Tell us what you'll do with 3 participants? 15? 25? *

With <10 participants we will be very hands on, cutting down the intro parts of the talk to spend more time on actual data, we will also work with those participants to see if they have existing data that is spatial that we can help then analyze.

With more we will have a more traditional lead process with prepared data and steps.

What do you see as outcomes after the festival? How will you and your participants take the learning and activities forward? In 150 words or less.

The biggest outcome for the session will be an awareness of the power of location intelligence and spatial approaches to data analysis. I would hope that participants will become aware of how much of their data can be interpreted in this way and the tools and services that are available to make that an easier process

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment