Click here to return to the home page.
This page contains vocabularies and taxonomies for classifying disaster types – they include the GLIDE vocabulary , the IFRC Disaster Types Taxonomy and the EM-DAT Disaster Types Taxonomy. When looking at the mindmaps, you will see that each term to the left Is Broader Than the terms to the right. In other words, all terms further down the tree can be aggregated up to the broader term. Conversely, reading from the right to the left, each term Is Narrower Than, or Is a Type of the term to its left. Click on the circles to expand/hide branches of the mindmap, and hover over terms to see more definitions.
In addition to relating terms within a hierarchy, taxonomies also allow us to define relationships between terms in different taxonomies. This gives humans and computers greater power to integrate and explore information.
Imagine comparing a dataset submitted to HDX by ‘Organisation 1’ using the GLIDE Disaster Type Terms, and another submitted to HDX by ‘Organisation 2’ using the IFRC Disaster Type Terms. Using the Broader/Narrower/Exact Match taxonomical terms we can pre-define the relationships – e.g. GLIDE Term Epidemic Is an Exact Match with IFRC Term Epidemic. IFRC Term Ebola Is a Narrower term than IFRC TERM Epidemic. With relationships pre-defined in this way, we can produce Taxonomy Translation Tools to automate the translation of terms, making combined analysis of datasets much easier and more meaningful.
WARNING: The Taxonomy Tool is only in DEMO phase at the moment. The Google Spreadsheet URL on the first page is an example of a dataset using GLIDE Disaster Type terms. Click Next and highlight the column with Disaster Type Terms. Click Next and select ‘disaster_GLIDE’ as the taxonomy to map from. Click Next and select ‘disaster_IFRC L4’ as the taxonomy to map to. Select relevant terms for the few items that are not auto-mapped.
Figure 1 – A screenshot of the translation tool showing all Narrower terms in the IFRC taxonomy for ‘Epidemic’.
Figure 2 – Combined analyses are more meaningful when terms are common between datasets. This is all made possible by the use of vocabularies and taxonomies.