It Makes Sense: A Wide-Coverage Word Sense Disambiguation System for Free Text

By Zhi Zhong, Hwee Tou Ng. Annual Meeting of the Association for Computational Linguistics, 2010

354 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/9931b8ea6594b97c7dfca93936a2d95a38167046

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Unsupervised Does Not Mean Uninterpretable: The Case for Word Sense Induction and Disambiguation (2017, background): “The IMS system by Zhong and Ng (2010) provides an implementation of the supervised approach to WSD that yields state-of-the-art results.”
AutoExtend: Extending Word Embeddings to Embeddings for Synsets and Lexemes (2015, methodology): “From the deﬁnition of E ( i,j ) , we derived that E ∈ R | S |× n ×| W |× n is normalized over the ﬁrst dimension (Eq.”
From Word to Sense Embeddings: A Survey on Vector Representations of Meaning (2018, unknown): “Methods to perform WSD can be roughly divided into two classes: supervised (Zhong & Ng, 2010; Iacobacci, Pilehvar, & Navigli, 2016; Yuan, Richardson, Doherty, Evans, & Altendorf, 2016; Raganato, Delli Bovi, & Navigli, 2017b; Luo, Liu, Xia, Chang, & Sui, 2018) and knowledge-based (Lesk, 1986;…”
Entity Linking meets Word Sense Disambiguation: a Unified Approach (2014, background): “However, supervised approaches require huge amounts of annotated data (Zhong and Ng, 2010; Shen et al., 2013; Pilehvar and Navigli, 2014), an effort which cannot easily be repeated for new domains and languages, while unsupervised ones suffer from data sparsity and an intrinsic difﬁ-culty in their…”
CoNLL-2012 Shared Task: Modeling Multilingual Unrestricted Coreference in OntoNotes (2012, methodology): “Word Sense This year we used the IMS (It Makes Sense) (Zhong and Ng, 2010) word sense tagger.20 Word sense information, unlike syntactic parse information is not central to approaches taken by current coreference systems and so we decided to use a better word sense tagger to get a good state of the ”
Embeddings for Word Sense Disambiguation: An Evaluation Study (2016, methodology): “Since we used the same system conﬁguration across the two runs, we conclude that the OMSTI corpus is either substantially smaller or less representative than the corpus used by Zhong and Ng (2010) for building the pre-trained models of IMS.”
Word Sense Disambiguation: A Unified Evaluation Framework and Empirical Comparison (2017, methodology): “• IMS+embeddings (Taghipour and Ng, 2015b; Rothe and Sch¨utze, 2015; Iacobacci et al., 2016).”
Neural Sequence Learning Models for Word Sense Disambiguation (2017, methodology): “As supervised systems, we considered Context2Vec (Melamud et al., 2016) and It Makes Sense (Zhong and Ng, 2010, IMS ), both the original implementation and the best conﬁguration reported by Iacobacci et al. (2016, IMS + emb ), which also integrates word embeddings using exponential decay.”
Semi-supervised New Event Type Induction and Event Detection (2020, methodology): “Speciﬁcally, for each word, we disambiguate its senses and link each sense to OntoNotes (Hovy et al., 2006) using a word sense disambiguation system — IMS (Zhong and Ng, 2010) 2 .”
Embedding Words and Senses Together via Joint Knowledge-Enhanced Training (2016, background): “tems (Zhong and Ng, 2010; Iacobacci et al., 2016; Melamud et al., 2016), could become an obstacle when applied to large corpora.”
Movie Description (2016, methodology): “…order to provide the linking of words in the sentence to their WordNet sense take off drink mappings, we rely on a state-of-the-art WSD system, IMS (Zhong and Ng 2010). give get In the following we refer to these as text-and sense-labels . eat Thus from each sentence we extract a semantic…”
Towards a Seamless Integration of Word Senses into Downstream NLP Applications (2017, methodology): “Moreover, unlike supervised WSD and entity linking techniques (Zhong and Ng, 2010; Cheng and Roth, 2013; Melamud et al., 2016; Limsopatham and Collier, 2016), our algorithm relies only on semantic networks and does not require any senseannotated data, which is limited to English and almost…”