ExaCT is an Information Extraction system developed jointly by the National Research Council of Canada and the University of California, San Francisco. The system is trained to find key information from scientific clinical trial publications, namely the descriptions of the trial's interventions, population, outcome measures, funding sources, and other critical characteristics. Automatically extracted answers are presented to a user through a web browser-based interface that assists the users in verifying and changing the extracted information. The video below demonstrates the main features of ExaCT's user interface.
The ExaCTDemo web site allows visitors to test the system's performance on their own data. To use the system, you have to create an account consisting of a user name and a password. After that, you will be able to upload and process articles on clinical trials and amend the automatic extractions. Only HTML files, single or ZIP archived, are currently accepted. For each user, a separate space is allocated on the server for all uploaded files. The files are visible only to the user who uploaded them to the server. Files can be deleted from the server by the user at any time. The User Guide provides further details on the interface. Also, the system has Help summaries accessible from the bottom part of each page.
© 2008-2010 University of California. Last modified 5-Jul-13