The Boundaries of Natural Language Processing Techniques in Extracting Knowledge from Emails

摘要

—The aim of this research is to determine if natural language processing techniques can be used to fully automate the extraction of knowledge from emails. The paper reviews the five generations of building systems to share knowledge and highlights the challenges faced by all. The paper discuss the system built by the authors and shows that although the f-measure results are world leading, there is still a requirement for user intervention to enable the system to be accurate enough to be of use to an organisation.