Evaluating WordNet-based Measures of Lexical Semantic Relatedness

By Alexander Budanitsky, Graeme Hirst. International Conference on Computational Logic, 2006

1,609 citations recorded by the source corpus.

The quantification of lexical semantic relatedness has many applications in NLP, and many different measures have been proposed. We evaluate five of these measures, all of which use WordNet as their central resource, by comparing their performance in detecting and correcting real-word spelling errors. An information-content-based measure proposed by Jiang and Conrath is found superior to those proposed by Hirst and St-Onge, Leacock and Chodorow, Lin, and Resnik. In addition, we explain why distributional similarity is not an adequate proxy for lexical semantic relatedness.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/0a32b3d027064798fb31ce42894fec31e834f7db

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

Combining Heterogeneous Models for Measuring Relational Similarity (2013, background): “Unlike the problem of attributional similarity, which measures whether two words share similar attributes and is addressed in extensive research work (Budanitsky and Hirst, 2006; Reisinger and Mooney, 2010; Radinsky et al., 2011; Agirre et al., 2009; Yih and Qazvinian, 2012), measuring relational si”
Multi-Prototype Vector-Space Models of Word Meaning (2010, background): “crafted dictionaries and cannot, in general, model sense frequency (Budanitsky and Hirst, 2006).”
Learning Entailment Relations by Global Graph Structure Optimization (2012, neutral): “WordNet (Fellbaum 1998b), by far the most widely used resource, speciﬁes relations such as hyponymy, synonymy, derivation , and entailment that can be used for semantic inference (Budanitsky and Hirst 2006).”
Argumentation mining: the detection, classification and structure of arguments in text (2009, background): “In the former, the relatedness of words depends on their semantic distances in a lexioo-semantic resource such as WordNet [5].”
A machine learning approach to sentiment analysis in multilingual Web texts (2009, background): “By comparing the semantic similarity (Budanitsky and Hirst 2004), i.e., the length of the path from a word to respectively the words “good” and “bad”, the semantic orientation of the word can be obtained.”
Addressing Age-Related Bias in Sentiment Analysis (2018, background): “Natural language processing techniques, including sentiment analysis, are a primary point for inquiry among both social scientists and machine learning researchers [11,18,33,51].”
SemEval-2017 Task 2: Multilingual and Cross-lingual Semantic Word Similarity (2017, methodology): “through the conventional closest senses strategy (Resnik, 1995; Budanitsky and Hirst, 2006), using cosine similarity as the comparison measure.”
KORE: keyphrase overlap relatedness for entity disambiguation (2012, background): “are a fundamental asset for many applications dealing with natural language processing (NLP), text mining, or Web analytics [4].”
Automatically Creating Datasets for Measures of Semantic Relatedness (2006, background): “The latter question is tackled by application-speciﬁc evaluation, where a measure is tested within the framework of a certain application, e.g. word sense disambiguation (Patwardhan et al., 2003) or malapropism detection (Budanitsky and Hirst, 2006).”
A Simple, Similarity-based Model for Selectional Preferences (2007, methodology): “The sim function can equally well be instantiated with a WordNet-based metric (for an overview see Budanitsky and Hirst (2006)), but we restrict our experiments to corpus-based metrics (a) in the interest of greatest possible resource-independence and (b) in order to be able to shape the similarity…”
Turkers, Scholars, "Arafat" and "Peace": Cultural Communities and Algorithmic Gold Standards (2015, background): “[32]) to high-level technologies like search [6, 34] and information visualization systems [4, 37, 3].”
Computing Semantic Relatedness Using Wikipedia-based Explicit Semantic Analysis (2007, methodology): “WordNet-based techniques, which only consider the generalization (“is-a”) relation between words, achieve correlation of only 0.33–0.35 with human judgements [Budanitsky and Hirst, 2006].”
Align, Disambiguate and Walk: A Unified Approach for Measuring Semantic Similarity (2013, neutral): “Approaches to semantic similarity have often operated at separate levels: methods for word similarity are rarely applied to documents or even single sentences (Budanitsky and Hirst, 2006; Radinsky et al., 2011; Halawi et al., 2012), while document-based similarity methods require more linguistic fea”