SURel: A Gold Standard for Incorporating Meaning Shifts into Term Extraction

Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM), pages 1–8 Minneapolis, June 6–7, 2019. c⃝2019 Association for Computational Linguistics 1 SURel: A Gold Standard for Incorporating Meaning Shifts into Term Extraction Anna H¨atty1,2, Dominik Schlechtweg2, Sabine Schulte im Walde2 1Robert Bosch GmbH 2Institute for Natural Language Proessing (IMS), University of Stuttgart anna.haetty@de.bosch.com, {schlecdk,schulte}@ims.uni-stuttgart.de Abstract We introduce SURel, a novel dataset for Ger- man with human-annotated meaning shifts be- tween general-language and domain-specific contexts. We show that meaning shifts of term candidates cause errors in term extraction, and demonstrate that the SURel annotation reflects these errors. Furthermore, we illustrate that SURel enables us to assess optimisations of term extraction techniques when incorporating meaning shifts. 1 Introduction Domain-specific terms often undergo meaning shifts from general-language use to their respec- tive domain-specific language use. For example, the German noun Schnee predominantly means ‘snow’ in its general-language usage, and ‘beaten egg whites’ in the cooking domain. Terms with these characteristics are referred to as sub- technical terms and pose a problem for term ex- traction: Their hybrid character makes it hard for humans to rank them along with unambiguous terms, and hard for computational models to clas- sify them as terms, because of the strong bias to- wards their general-language meanings. In this study, we present SURel (Synchronic Usage Relatedness), a novel dataset for meaning shifts from general to domain-specific language, based on human annotations on the degrees of se- mantic relatedness between contexts of term can- didates. We illustrate that SURel reflects the error that is commonly made by term extraction mea- sures for sub-technical terms when relying on a general-language reference corpus. In a first ex- periment, we predict the meaning shift automat- ically and use SURel for evaluation. We then in- corporate the model’s prediction as a factor into an established term extraction measure, to correct the error in termhood prediction caused by meaning shifts. 2 Meaning Shifts in Terminology Sub-Technical Terms Terms are linguistic units that characterize specialized domains (Kageura and Umino, 1996), thus representing opposite ex- tremes of words that are not specific to a domain (Sager, 1990). Sub-technical terms (Cowan, 1974; Trimble, 1985; Baker, 1988; Chung and Nation, 2003; P´erez, 2016) occupy intermediary positions on the continuum, because they undergo meaning shifts from general to domain-specific language usage. Baker (1988) distinguishes two types of sub-technical terms with general-language usage: words with a restricted domain-specific meaning (e.g., effective means ‘take effect’ in biology), and words with a complete meaning shift (e.g., bug in computer science). Sub-technical terms are a major problem for term extraction measures which often operate on the word type rather than the word sense level. P´erez (2016) provides empirical evidence that 50% of legal terminology is represented by sub- technical terms. Lay people often do not even no- tice their terminological character due to their pre- dominant general-language use (H¨atty and Schulte im Walde, 2018). Term Extraction Techniques One of the main strands of term extraction methodologies are con- trastive techniques, which …

Content truncated to 500 words