Automatic Retrieval and Clustering of Similar Words

By Dekang Lin. 1998

Bootstrapping semantics from text is one of the greatest challenges in natural language learning. We first define a word similarity measure based on the distributional pattern of words. The similarity measure allows us to construct a thesaurus using a parsed corpus. We then present a new evaluation methodology for the automatically constructed thesaurus. The evaluation results show that the thesaurus is significantly closer to WordNet than Roget Thesaurus is.

Subfields: Information Retrieval & Search, Natural Language Processing

How researchers see this paper

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

In Information Retrieval & Search

In Natural Language Processing

Citations in the corpus

Zero-Shot Learning Through Cross-Modal Transfer (2013, background): “These representations have proven very effective in natural language processing tasks such as sense disambiguation [30], thesaurus extraction [24, 8] and cognitive modeling [22].”
From Frequency to Meaning: Vector Space Models of Semantics (2010, background): “…similarity by inversion (13) or subtraction (14). sim(x,y) = 1/dist(x,y) (13) sim(x,y) = 1− dist(x,y) (14) Many similarity measures have been proposed in both IR (Jones & Furnas, 1987) and lexical semantics circles (Lin, 1998; Dagan, Lee, & Pereira, 1999; Lee, 1999; Weeds, Weir, & McCarthy, 2004).”
Dependency-Based Word Embeddings (2014, background): “Specifically, 1code.google.com/p/word2vec/ 302 following work in sparse vector-space models (Lin, 1998; Padó and Lapata, 2007; Baroni and Lenci, 2010), we experiment with syntactic contexts that are derived from automatically produced dependency parse-trees.”
Linguistic Regularities in Sparse and Explicit Word Representations (2014, background): “…been extensively studied in the NLP literature (see (Tur-ney and Pantel, 2010; Baroni and Lenci, 2010) and the references therein), and are known to exhibit a large extent of attributional similarity (Pereira et al., 1993; Lin, 1998; Lin and Pantel, 2001; Sahlgren, 2006; Kotlerman et al., 2010).”
A Structured Vector Space Model for Word Meaning in Context (2008, background): “This corresponds to a beneﬁcial effect of syntactic information found for other applications of semantic spaces (Lin, 1998; Pad ´ o and Lapata, 2007).”
Distributional Memory: A General Framework for Corpus-Based Semantics (2010, neutral): “The W 1 × LW 2 matrix is a structured semantic space similar to those used by Curran and Moens (2002), Grefenstette (1994), and Lin (1998a), among others.”
Mathematical Foundations for a Compositional Distributional Model of Meaning (2010, background): “Examples include lexicon acquisition [16, 26], word sense discrimination and disambiguation [36, 28], text segmentation [5], language modelling [2], and notably document retrieval [35].”
The New C Standard (Excerpted material) An Economic and Cultural Commentary (2008, unknown): “Lin[207] proposed a similarity measure that takes the grammatical relationship between the two words into account.”
Generating Phrasal and Sentential Paraphrases: A Survey of Data-Driven Methods (2010, background): “…work on purely lexical paraphrasing which usually comprises various ways to cluster words occurring in similar contexts (Inoue 1991; Crouch and Yang 1992; Pereira, Tishby, and Lee 1993; Grefenstette 1994; Lin 1998; Gasperin et al. 2001; Glickman and Dagan 2003; Shimohata and Sumita 2005).”
A Survey of Paraphrasing and Textual Entailment Methods (2009, background): “WordNet (Fellbaum, 1998), automatically constructed collections of near synonyms (Lin, 1998a; Moore, 2001; Brockett & Dolan, 2005), or resources like NOMLEX (Meyers, Macleod, Yangarber, Grishman, Barrett, & Reeves, 1998) and CATVAR (Habash & Dorr, 2003) that provide nominalizations of verbs and…”
Vector-based Models of Semantic Composition (2008, methodology): “Under this framework, we introduce novel composition models which we compare empirically against previous work using a rigorous evaluation methodology.”
A Confidence Model for Syntactically-Motivated Entailment Proofs (2011, methodology): “, 2009), Lin similarity (Lin, 1998a), and Directional-Similarity8 (Kotlerman et al.”
Modeling Newswire Events using Neural Networks for Anomaly Detection (2014, methodology): “This can be achieved by integrating real world knowledge from knowledge bases like Wordnet (Miller, 1995) or from corpus statistics like the work by Lin (1998) into the event model.”
A survey of named entity recognition and classification (2007, methodology): “A baseline SL method that is often proposed consists of tagging words of a test corpus when they are annotated as entities in the training corpus. The performance of the baseline system depends on the vocabulary transfer, which is the proportion of words, without repetitions, appearing in both train”
Polarity Inducing Latent Semantic Analysis (2012, methodology): “Lin et al. (2003) builds on (Lin, 1998) and identiﬁes antonyms as semantically related words which also happen to be found together in a database in pre-identiﬁed phrases indicating opposition.”
Measuring semantic similarity between words using web search engines (2007, methodology): “It is comparable with Lin (1998) [17] Jiang & Conrath (1998) [13] and Li (2003) [41] methods.”