WordNet::Similarity - Measuring the Relatedness of Concepts

By Ted Pedersen, Siddharth Patwardhan, Jason Michelizzi. North American Chapter of the Association for Computational Linguistics, 2004

1,854 citations recorded by the source corpus.

WordNet::Similarity is a freely available software package that makes it possible to measure the semantic similarity and relatedness between a pair of concepts (or synsets). It provides six measures of similarity, and three measures of relatedness, all of which are based on the lexical database WordNet. These measures are implemented as Perl modules which take as input two concepts, and return a numeric value that represents the degree to which they are similar or related.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/495f3405da229b903797472c64d09d83659fdb34

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Automated Crowdturfing Attacks and Defenses in Online Review Systems (2017, methodology): “Relevancy is estimated by calculating lexical similarity using WordNet [44], a widely used lexical database that groups English words into sets of synonyms and measures their concept relatedness [51].”
SimLex-999: Evaluating Semantic Models With (Genuine) Similarity Estimation (2014, methodology): “This is supported by the fact that the WordNet-based methods that perform best at modeling human ratings model the similarity between concepts c1 and c2 as the minimum of all pairwise distances between the senses of c1 and the senses of c2 (Resnik, 1995; Pedersen et al., 2004).”
Composition in Distributional Models of Semantics (2010, methodology): “We used the implementation provided in the WordNet Similarity package (Pedersen, Patwardhan, & Michelizzi, 2004).”
Matching and Merging of Statecharts Specifications (2007, methodology): “We employ the freely available WordNet::Similarity package [23] for this purpose.”
Predicting Tasks in Goal-Oriented Spoken Dialog Systems using Semantic Knowledge Bases (2013, methodology): “Semantic Similarity between two terms can be computed using different metrics (Pedersen et al., 2004) based on resources like WordNet.”
A New Set of Norms for Semantic Relatedness Measures (2013, methodology): “…WordNet-based measures from Table 2 that are regarded in the literature as relatedness measures (Banerjee and Pedersen, 2003; Hirst and St-Onge, 1998; Patwardhan and Pedersen, 2006) 5Results based on standard implementations in the WordNet::Similarity Perl module of Pedersen et al. (2004) (v2.”
Ontology learning from text: A look back and into the future (2012, background): “1998], word sense disambiguation [Vronis and Ide 1998; Lesk 1986], as well as similarity measurement [Pedersen et al. 2004].”
Knowledge Graphs: Opportunities and Challenges (2023, background): “The knowledge base contains a number of hierarchical concept graphs to analyse the semantic similarity (Pedersen et al. 2004).”
Re-evaluating Automatic Metrics for Image Captioning (2016, methodology): “For tuple matching, SPICE uses WordNet synonym matching (Pedersen et al., 2004) as in METEOR (Banerjee and Lavie, 2005).”
Evaluating WordNet-based Measures of Lexical Semantic Relatedness (2006, background): “Since the ﬁrst publication of the initial results of this work (Budanitsky and Hirst 2001), Pedersen and his colleagues (Pedersen, Patwardhan, and Michelizzi 2004) have made available a Perl implementation of the ﬁve WordNet-based measures (plus Wu and Palmer’s and their own; see below) that has…”
A latent factor model for highly multi-relational data (2012, methodology): “’s word embeddings [5] and the best (out of 3) WordNet Similarity measure [19] using area under the precision-recall curve.”
Relational Generalized Few-Shot Learning (2019, methodology): “In particular, we use WordNet path similarities [39] between class labels, which are scores based on the shortest path distances between words in the taxonomy.”
Soft Cardinality: A Parameterized Similarity Function for Text Comparison (2012, background): “Furthermore, hybrid approaches can integrate similarity functions that are not based on the representation of the surface of text, such as semantic relatedness measures (Pedersen et al., 2004).”
The Eras and Trends of Automatic Short Answer Grading (2015, background): “In Auto-Assessor, each word that matches exactly is given one point, related words from WordNet are given partial credit, and the rest are given no credit.”
Going Beyond Text: A Hybrid Image-Text Approach for Measuring Word Relatedness (2011, methodology): “We evaluate this metric alongside existing corpus-based (Tur-ney and Pantel, 2010) and knowledge-based metrics (Pedersen et al., 2004) either in a standalone or combined setting and present our ﬁndings.”
Structured Lexical Similarity via Convolution Kernels on Dependency Trees (2011, background): “…of generalized features or the deﬁnition of semantic similarities between them, e.g. as proposed in (Resnik, 1995; Jiang and Conrath, 1997; Schtze, 1998; Pedersen et al., 2004a; Bloehdorn and Mos-chitti, 2007b; Davis et al., 2007) or in semi-supervised settings, e.g. (Chapelle et al., 2006).”
Directional distributional similarity for lexical inference (2010, methodology): “…amongst others the following competitive measures: Lch by Leacock and Chodorow (1998), Lesk by Baner-jee and Pedersen (2002), Lin by Lin (1998c), Path by Pedersen et al. (2004), Res by Resnik (1995), Vector by Patwardhan (2003), Wup by Wu and Palmer (1994), and Jcn by Jiang and Conrath (1997).”
Eviza: A Natural Language Interface for Visual Analysis (2016, background): “unit taxonomy [9] and Wordnet’s synsets [29].”