Concept-based Mining to Enhance the Scope and Speed of Archival Qualitative Research
Andrew E. Smith
Key Centre for Human Factors and Applied Cognitive Psychology
The University of Queensland Queensland, Australia, 4072.
Email address of corresponding author: asmith@humanfactors.uq.edu.au
The development of a pilot system for archiving and retrieval of qualitative data in Australia is described. This system, called the Australian Qualitative Archive (AQuA), is not only intended as a valuable historical record, but as a key element in allowing qualitative research to achieve greater scale and impact through data sharing and secondary analysis. Limitations in existing retrieval techniques such as static classification and keyword search are discussed, and some solutions are suggested involving the Leximancer text analysis system. The functional architecture of the archive is described, including: ways to store the context and method descriptions alongside the data; ways to store coding and annotation analysis product as well as original data; ways for summary data to be retrieved when access to raw data is restricted; ways for qualitative data to be browsed and analysed on-line using emergent classification systems, or concept maps, extracted using Leximancer.
