Automatic subject indexing and evaluation: an introduction
In this talk Koraljka Golub introduced basic principles of automating the process of subject classification and indexing, with a focus on text. Different approaches such as text categorization, document clustering and document classification were presented, compared for purposes of application, with most basic principles of automation introduced without going into advanced computational detail. Complexities of evaluation were also addressed.
The talk was largely based on the following open access encyclopedia article: Golub, Koraljka. 2019. "Automatic subject indexing of text". Knowledge Organization 46, no. 2: 104-121. Also available in ISKO Encyclopedia of Knowledge Organization, eds. Birger Hjørland and Claudio Gnoli, http://www.isko.org/cyclo/automatic.
The class was delivered via Zoom and was attended by 88 participants from 33 countries. Presentation slides are available to all participants via the KO-ED forum on the ISKO UK website.