Tutorial: using aurin's geocoder chooser tool

Introduction

As part of the first phase of the Tinker project (information about AURIN’s involvment can be found here), AURIN developed a geo-coder chooser tool for Digital Humanities Scholars. The purpose of the tool is to allow Digital Humanities Scholars to make sense of the geocoder options that exist, and to hopefully find the right tool to fit their data and analytical needs.

What is Geocoding?

What is Geocoding?

Geocoding is the process by which a description of a place, whether that be the name of a location, or an address, is converted to a point or a place in space, most often represented by a latitude and a longtitude. You are probably familiar with a very common geocoder, Google Maps, where, if you enter a street address into a search bar, it will specify a point in space (as shown below, for arguably the most famous street address in the world)



Alternatively, you can use the same “geocoder” to enter the name of a location, rather than a street address, and it will often provide you with a point in space, such as AURIN’s host building at the University of Melbourne. 



Athough the latitude and longitude of these points are not shown by default, it is possible to get the latitude and longitude of points in this particular geocoder selecting the “What’s Here” option for any point in space:



You can also enter the name of a place that is larger than a single point – geocoding involves more than points, and it is possible to geocode to an area or a defined boundary (shown below), rather than a specific point location. This is often a better solution than going to a point location if, for example, you don’t have more information than a suburb or neighbourhood – gecoding to a point location would involve spurious precision



Of course, it’s not practical to manually enter the addresses of a database with potentially thousands of entries, so you need an automated tool that will do it for you, taking a database and then producing a list of point locations. Luckily there are plenty of those out there, but they each have different “fits for purpose”. This tool provides some guidance in choosing the right geocoder

Using the Tinker Geocoder Chooser

Using the Tinker Geocoder Chooser

  • Go to: https://tinker.edu.au/tools/geo-chooser/


  • Enter the details of your datasets and your needs – it is very important that you understand your data structure first before using the chooser. Important questions are:
  • Does your dataset contain descriptions of places, or actual addresses? A mixture of both?
  • Are you addresses spread out over multiple fields, or in one field separated by columns e.g.
Number Street Suburb State Postcode
123 Street Road Suburbia Victoria 3000

vs

Address
123 Street Road, Suburbia, Victoria, 3000
  • Are you looking at world wide addresses? Or Just Australia?
  • Are you only looking at current addresses? Or do they include historical addresses? A mixture of both?
  • What are the outputs you want? Point locations or areas?

A worked example is shown below, where the user has specified that:

  1. They require a stand-alone tool
  2. The dataset only contains addresses
  3. The data are structured so that the address is spread over multiple fields
  4. The user is only looking at Australia
  5. The data are current, rather than historical
  6. The output requires a point location

As a result of this, several geocoders are nominated as fit for purpose