Similarity Metrics within a Point of View

Vector space based approaches to natural language processing are contrasted with human similarity judgements to show the manner in which human subjects fail to produce data which satisfies all requirements for a metric space. This result would constrains the validity and applicability vector space based (and hence also quantum inspired) approaches to the modelling of cognitive processes. This paper proposes a resolution to this problem, by arguing that pairs of words imply a context which in turn induces a point of view, so allowing a subject to estimate semantic similarity. Context is here introduced as a point of view vector (POVV) and the expected similarity is derived as a measure over the POVV's. Different pairs of words will invoke different contexts and different POVV's. We illustrate the proposal on a few triples of words and outline further research.

[1]  Patrick Pantel,et al.  From Frequency to Meaning: Vector Space Models of Semantics , 2010, J. Artif. Intell. Res..

[2]  Diederik Aerts,et al.  Quantum Structure in Cognition , 2008, 0805.3850.

[3]  A. Tversky,et al.  Similarity, separability, and the triangle inequality. , 1982, Psychological review.

[4]  J. Firth,et al.  Papers in linguistics, 1934-1951 , 1957 .

[5]  Richard A. Harshman,et al.  Indexing by Latent Semantic Analysis , 1990, J. Am. Soc. Inf. Sci..

[6]  Jordan L. Boyd-Graber,et al.  Adding dense, weighted connections to WordNet , 2005 .

[7]  T. Landauer,et al.  A Solution to Plato's Problem: The Latent Semantic Analysis Theory of Acquisition, Induction, and Representation of Knowledge. , 1997 .

[8]  A. Tversky.  Features of Similarity , 1977 .

[9]  R. Nosofsky.  Attention, similarity, and the identification-categorization relationship. , 1986, Journal of experimental psychology. General.

[10]  Peter Bruza,et al.  Is There Something Quantum-Like about the Human Mental Lexicon? , 2009, INEX.

[11]  Diederik Aerts,et al.  A Theory of Concepts and Their Combinations II: A Hilbert Space Representation , 2004 .

[12]  J. Firth.  Papers in linguistics , 1958 .

[13]  Hinrich Schütze,et al.  Automatic Word Sense Discrimination , 1998, Comput. Linguistics.

[14]  Michael N Jones,et al.  Representing word meaning and order information in a composite holographic lexicon. , 2007, Psychological review.

[15]  J. Bullinaria,et al.  Extracting semantic representations from word co-occurrence statistics: A computational study , 2007, Behavior research methods.

[16]  Hinrich Sch.  Automatic Word Sense Discrimination , 1998 .

[17]  E. Rosch.  Cognitive Representations of Semantic Categories. , 1975 .

[18]  Diederik Aerts,et al.  A theory of concepts and their combinations I: The structure of the sets of contexts and properties , 2005 .

[19]  William K. Wootters.  The Acquisition of Information from Quantum Measurements. , 1980 .

[20]  Vladislav D. Veksler,et al.  Defining the Dimensions of the Human Semantic Space , 2008 .

[21]  P. Kanerva,et al.  Permutations as a means to encode order in word space , 2008 .

[22]  Lance J. Rips,et al.  Combining Prototypes: A Selective Modification Model , 1988, Cogn. Sci..

[23]  Gerard Salton,et al.  A vector space model for automatic indexing , 1975, CACM.

[24]  Alfred Inselberg.  Visualization of concept formation and learning , 2005 .

[25]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[26]  Laurianne Sitbon,et al.  Quantum-like non-separability of concept combinations, emergent associates and abduction , 2012, Log. J. IGPL.

[27]  Zellig S. Harris,et al.  Distributional Structure , 1954 .

[28]  E. Rosch,et al.  Cognition and Categorization , 1980 .

[29]  Magnus Sahlgren,et al.  An Introduction to Random Indexing , 2005 .