Improving the Coverage and the Generalization Ability of Neural Word Sense Disambiguation through Hypernymy and Hyponymy Relationships

2 Nov 2018Loïc VialBenjamin LecouteuxDidier Schwab

In Word Sense Disambiguation (WSD), the predominant approach generally involves a supervised system trained on sense annotated corpora. The limited quantity of such corpora however restricts the coverage and the performance of these systems... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Word Sense Disambiguation SemEval 2007 Task 17 SemCor+WNGT, vocabulary reduced, ensemble F1 66.81 # 2
Word Sense Disambiguation SemEval 2007 Task 7 SemCor+WNGT, vocabulary reduced, ensemble F1 86.02 # 2
Word Sense Disambiguation SemEval 2013 Task 12 SemCor+WNGT, vocabulary reduced, ensemble F1 72.63 # 2
Word Sense Disambiguation SemEval 2015 Task 13 SemCor+WNGT, vocabulary reduced, ensemble F1 74.46 # 2
Word Sense Disambiguation SensEval 2 SemCor+WNGT, vocabulary reduced, ensemble F1 75.15 # 2
Word Sense Disambiguation SensEval 3 Task 1 SemCor+WNGT, vocabulary reduced, ensemble F1 70.11 # 7
Word Sense Disambiguation Supervised: SemCor+WNGT, vocabulary reduced, ensemble Senseval 2 75.15 # 2
Word Sense Disambiguation Supervised: SemCor+WNGT, vocabulary reduced, ensemble Senseval 3 70.11 # 5
Word Sense Disambiguation Supervised: SemCor+WNGT, vocabulary reduced, ensemble SemEval 2007 66.81 # 2
Word Sense Disambiguation Supervised: SemCor+WNGT, vocabulary reduced, ensemble SemEval 2013 72.63 # 2
Word Sense Disambiguation Supervised: SemCor+WNGT, vocabulary reduced, ensemble SemEval 2015 74.46 # 2