XML version of Edinburgh Associative Thesaurus (EAT)

The Edinburgh Associative Thesaurus (EAT) is a set of word association norms showing the counts of word association as collected from subjects. This is not a developed semantic network such as WordNet, but empirical association data.

On this page, there is an interactive web form to consult the associations between a stimulus and a response. We can also download a zip archive of the files and C programs to use them in applications.

Unfortunately, the programs are a bit old and need some adaptations to run on modern machines and compilers. But the stimulus-response data (sr_concise) is in a relatively user-friendly format.

In order to make it more easily tractable by modern tools, I have produced from the stimulus-response data, an XML file that gives for each stimulus, the corresponding responses. From that file, I produced another XML that give for each response, the stimuli that produced them.

See this page, for an explanation of the numbers in the file. The match the ones that can be obtained by requests on the web page.

eat-stimulus-response.xml.zipA zipped archive [1.6M] of the stimulus response file, the stimuli are sorted in descending order first by total number of responses and then by the number of different responses. Once unzipped, the file is 16 MEGs.
eat-stimulus-response.xslThe XSLT stylesheet to produce the XML file from the rs_concise file.
eat-response-stimulus.xml.zipA zipped archive [1.5M] of the response stimulus file, the responses are sorted in descending order first by total number of stimuli and then by the number of different stimuli. Once unzipped, the file is 14 MEGs.
eat-response-stimulus.xslThe XSLT stylesheet to produce the XML file from the eat-stimulus-response.xml file.
eat100.xmlA small excerpt of the stimulus-response file, showing the 10 most frequent responses for the 100 most frequent stimuli.

For a similar type of information from another source see the XML version of the University of South Florida Free Association Norms

For some simple exploration and exploration of large XML files, you might like to look at this page

 

Guy Lapalme (June 2010)