What is Unithood

As more electronic text is readily available, and more applications become knowledge intensive and ontology-enabled, term extraction, also known as automatic term recognition or terminology mining is increasingly in demand. This chapter first presents a comprehensive review of the existing techniques, discusses several issues and open problems that prevent such techniques from being practical in real-life applications, and then proposes solutions to address these issues. Keeping afresh with the recent advances in related areas such as text mining, we propose new measures for the determination of unithood, and a new scoring and ranking scheme for measuring termhood to recognise domain-specific terms. The chapter concludes with experiments to demonstrate the advantages of our new approach.