The Word-Space Model: using distributional analysis to represent syntagmatic and paradigmatic relations between words in high-dimensional vector spaces

By Magnus Sahlgren. 2006

602 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/4996835ae239edc4d213940bfefa6d2c7e0ffbcd

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

Domain and Function: A Dual-Space Model of Semantic Relations and Compositions (2012, background): “There are many kinds of word–context matrices, based on various notions of context; Sahlgren (2006) gives a good overview of the types of context that have been explored in past work.”
Issues in evaluating semantic spaces using word analogies (2016, result): “The advantage of smaller window sizes in capturing “syntactic” information is consistent with previous studies (Redington et al., 1998; Sahlgren, 2006).”
Human and Machine Judgements for Russian Semantic Relatedness (2016, methodology): “Finally, Sahlgren [28] evaluated distributional lexical similarity measures comparing them to manually-crafted thesauri, but also associative norms, such as those described in the following section.”
Text: now in 2D! A framework for lexical expansion with contextual similarity (2013, background): “Magnus Sahlgren (2006), The Word-Space Model: using distributional analysis to represent syntagmatic and paradigmatic relations between words in high-dimensional vector spaces.”
Don’t count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors (2014, background): “…of alternative count DSMs on standard benchmarks (Agirre et al., 2009; Baroni and Lenci, 2010; Bullinaria and Levy, 2007; Bullinaria and Levy, 2012; Sahlgren, 2006; Padó and Lapata, 2007), the existing literature contains very little in terms of direct comparison of count vs. predictive DSMs.”
A Compositional and Interpretable Semantic Space (2015, background): “VSMs have been shown to closely match human judgements of semantics (for an overview see Sahlgren (2006), Chapter 5), and can be used to study semantic composition (Mitchell and Lapata, 2010; Baroni and Zamparelli, 2010; Socher et al., 2012; Turney, 2012).”
Word Representations: A Simple and General Method for Semi-Supervised Learning (2010, methodology): “Sahlgren (2006) does a battery of experiments exploring di ﬀ erent design decisions involved in constructing F , prior to using random indexing.”
Structured Lexical Similarity via Convolution Kernels on Dependency Trees (2011, background): “The contexts are words appearing in a n -window with target words: such a space models a generic notion of semantic relatedness, i.e. two words close in the space are likely to be either in paradigmatic or syntagmatic relation as in (Sahlgren, 2006).”
Dependency-Based Construction of Semantic Space Models (2007, methodology): “Sahlgren (2006) uses Random Indexing, a method comparable to LSA, to represent the meaning of words and reports a 75.0% accuracy on the same TOEFL items.”
WordICA—emergence of linguistic representations for words by independent component analysis (2010, methodology): “This diﬀerence in the eﬀect of using short versus long contexts is highlighted by comparing the results of earlier experiments using diﬀerent methods (see, e.g., Bingham et al. 2001, 2002; Levy, Bullinaria and Patel 1998; Sahlgren 2006) and the results reported later in this paper.”
Learning word embeddings efficiently with noise-contrastive estimation (2013, methodology): “Some classic word-space models such as HAL and COALS [16] follow this approach by representing the context distribution using a bag-of-words but they do not learn embeddings from this information.”
Word Embeddings Go to Italy: A Comparison of Models and Training Datasets (2015, background): “Research on word representation models [19,20], word embeddings, has gained a lot of attention in the recent years [9].”
Semantic Vector Products: Some Initial Investigations (2008, background): “Semantic vector models are by now a recognized part of mainstream computational linguistics , and are sometimes described as wordspace models (Widdows 2004; Sahlgren 2006).”
Linguistic Distributional Knowledge and Sensorimotor Grounding both Contribute to Semantic Category Production (2020, background): “…that are more frequent than expected.4 Linguistic proximity is therefore a measure of first-order co-occurrence, reflecting the extent to which two words co-occur more often than chance, which can capture a complex range of semantic relations (Sahlgren, 2006; Wingfield & Connell, 2020).”