Adverbs Revisited: Enhancing WordNet Coverage of Adverbs with a Supersense Taxonomy
This work addresses a gap in semantic resources for NLP applications, offering incremental improvements to WordNet's coverage for tasks like word sense disambiguation and sentiment analysis.
The authors tackled the problem of WordNet's underdeveloped adverb coverage by introducing a linguistically grounded supersense typology for adverbs, which was empirically validated through annotation to provide broad coverage and reliable assignment in natural text.
WordNet offers rich supersense hierarchies for nouns and verbs, yet adverbs remain underdeveloped, lacking a systematic semantic classification. We introduce a linguistically grounded supersense typology for adverbs, empirically validated through annotation, that captures major semantic domains including manner, temporal, frequency, degree, domain, speaker-oriented, and subject-oriented functions. Results from a pilot annotation study demonstrate that these categories provide broad coverage of adverbs in natural text and can be reliably assigned by human annotators. Incorporating this typology extends WordNet's coverage, aligns it more closely with linguistic theory, and facilitates downstream NLP applications such as word sense disambiguation, event extraction, sentiment analysis, and discourse modeling. We present the proposed supersense categories, annotation outcomes, and directions for future work.