Comprehensive Supersense Disambiguation of English Prepositions and Possessives

Semantic relations are often signaled with prepositional or possessive marking--but extreme polysemy bedevils their analysis and automatic interpretation. We introduce a new annotation scheme, corpus, and task for the disambiguation of prepositions and possessives in English. Unlike previous approaches, our annotations are comprehensive with respect to types and tokens of these markers; use broadly applicable supersense classes rather than fine-grained dictionary definitions; unite prepositions and possessives under the same class inventory; and distinguish between a marker's lexical contribution and the role it marks in the context of a predicate or scene. Strong interannotator agreement rates, as well as encouraging disambiguation results with established supervised methods, speak to the viability of the scheme and task.

PDF Abstract ACL 2018 PDF ACL 2018 Abstract

Results from the Paper


Ranked #4 on Natural Language Understanding on STREUSLE (Role F1 (Preps) metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Natural Language Understanding STREUSLE BiLSTM + MLP (gold syntax) Role F1 (Preps) 62.2 # 4
Function F1 (Preps) 73.4 # 4
Full F1 (Preps) 58.9 # 7
Natural Language Understanding STREUSLE BiLSTM + MLP (auto syntax) Role F1 (Preps) 56.3 # 7
Function F1 (Preps) 66.7 # 6
Full F1 (Preps) 53.6 # 11
Natural Language Understanding STREUSLE SVM (feature-rich, auto syntax) Role F1 (Preps) 58.2 # 6
Function F1 (Preps) 66.7 # 6
Full F1 (Preps) 55.7 # 10
Natural Language Understanding STREUSLE SVM (feature-rich, gold syntax) Role F1 (Preps) 62.2 # 4
Function F1 (Preps) 71.0 # 5
Full F1 (Preps) 59.5 # 6

Methods


No methods listed for this paper. Add relevant methods here