The industry client is the Minerals Research Institute of WA (MRIWA). MRIWA has a database of pdf reports and will make a selection available for this project. The industry mentor is Nicole Rooke, the CEO of MRIWA . Nicole will be supported by Prof. Melinda Hodkiewicz (Engineering) and Caitlin Woods (CSSE), both from the NLP-TLP group at UWA.
The goals of this project are to 1) select suitable open source tool(s), 2) develop a pipeline to answer competency questions on a data set of pdfs stored on the MRIWA SharePoint site with a text-based answer to a question (as with ChatGPT), 3) provide references to the pdfs from which the response is drawn, 4) report on lessons learned and gaps, 5) provide concrete suggestions for next steps, and 6) provide access to the code in GitHub. MRIWA will provide a set of competency questions.
This is a meaningful and timely industry-relevant project as many organisations are seeking to use developments demonstrated by ChatGPT to change the way they can interogate their internal documents.
Department of Computer Science & Software Engineering The University of Western Australia Last modified: 12 July 2023 Modified By: Michael Wise |