Dimensions of meaning

By Hinrich Schütze. 1992

Subfields: Natural Language Processing, Computational Linguistics & Semantics

How researchers see this paper

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

In Natural Language Processing

In Computational Linguistics & Semantics

Citations in the corpus

Enriching Word Vectors with Subword Information (2016, methodology): “These representations are typically derived from large unlabeled corpora using co-occurrence statistics (Deerwester et al., 1990; Schütze, 1992; Lund and Burgess, 1996).”
Bag of Tricks for Efficient Text Classification (2016, background): “Common solutions to this problem are to factorize the linear clas-siﬁer into low rank matrices (Schutze, 1992; Mikolov et al., 2013) or to use multilayer neural networks (Collobert and Weston, 2008; Zhang et al., 2015).”
Unsupervised learning for text-to-speech synthesis (2013, methodology): “] the vector representations have the key properties of the distributed representations characteristic of parallel distributed processing (Schütze, 1992).”
Detect & Describe: Deep Learning of Bank Stress in the News (2015, background): “Modeling of word contexts yields distributed representations of word semantics as vectors, which allow measuring of semantic similarities and detecting analogies without supervision, given substantial amounts of text [18], [19], [9].”
Bank distress in the news: Describing events through deep learning (2016, background): “Word contexts are modeled to yield distributed representations of word semantics as vectors, as opposed to declarative formats, which allow measuring of semantic similarities and detecting analogies without supervision, given substantial amounts of text [35,36,24].”
Verb similarity on the taxonomy of WordNet (2006, methodology): “Word similarity is established by comparing distance measures such as the cosine coefficient or Euclidean distance (Schütze, 1992).”
Integrating experiential and distributional data to learn semantic representations. (2009, background): “In this respect, one of the first attempts to adequately address this hypothesis was due to Schütze (1992).”
Toward a theory of semantic representation (2009, methodology): “…in computational cognitive science, we are assuming here that important aspects of meaning representation are learnt on the basis of the statistical distribution of words across texts (Schutze 1992; Landauer and Dumais 1997; Burgess and Lund 1997; Lund and Burgess 1996; Gri‰ths et al. 2007).”
A Solution to Plato's Problem: The Latent Semantic Analysis Theory of Acquisition, Induction, and Representation of Knowledge. (1997, unknown): “Indeed, Lund and Burgess (in press; Lund, Burgess, & Atchley, 1995), and SchUtze (1992a, 1992b), have explored related models in which such data are the input.”
Can Language Models Encode Perceptual Structure Without Grounding? A Case Study in Color (2021, background): “Distributional word representations have long been theorized to capture various types of information about the world (Schütze, 1992).”
WordICA—emergence of linguistic representations for words by independent component analysis (2010, methodology): “Sch¨utze (1992) has developed the Word Space method that utilizes both word-word matrices and SVD for representing word meaning and senses.”