Don’t count, predict! A systematic comparison of context-counting vs. context-predicting semantic vectors

Context-predicting models (more commonly known as embeddings or neural language models) are the new kids on the distributional semantics block. Despite the buzz surrounding these models, the literature is still lacking a systematic comparison of the predictive models with classic, count-vector-based distributional semantic approaches. In this paper, we perform such an extensive evaluation, on a wide range of lexical semantics tasks and across many parameter settings. The results, to our own surprise, show that the buzz is fully justified, as the context-predicting models obtain a thorough and resounding victory against their count-based counterparts.

[1]  John B. Goodenough,et al.  Contextual correlates of synonymy , 1965, CACM.

[2]  Gene H. Golub,et al.  Matrix computations , 1983 .

[3]  G. Miller,et al.  Contextual correlates of semantic similarity , 1991 .

[4]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .

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

[6]  M. Tanenhaus,et al.  Modeling the Influence of Thematic Fit (and Other Constraints) in On-line Sentence Comprehension , 1998 .

[7]  Yoshua Bengio,et al.  A Neural Probabilistic Language Model , 2003, J. Mach. Learn. Res..

[8]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[9]  George Karypis,et al.  CLUTO - A Clustering Toolkit , 2002 .

[10]  Ehud Rivlin,et al.  Placing search in context: the concept revisited , 2002, TOIS.

[11]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[12]  Stefan Evert,et al.  The Statistics of Word Cooccur-rences: Word Pairs and Collocations , 2004 .

[13]  Magnus Sahlgren,et al.  The Word-Space Model: using distributional analysis to represent syntagmatic and paradigmatic relations between words in high-dimensional vector spaces , 2006 .

[14]  Abdulrahman Almuhareb,et al.  Attributes in lexical acquisition , 2006 .

[15]  Ulrike Padó,et al.  The integration of syntax and semantic plausibility in a wide-coverage model of human sentence processing , 2007 .

[16]  Mark Steyvers,et al.  Topics in semantic representation. , 2007, Psychological review.

[17]  Mirella Lapata,et al.  Dependency-Based Construction of Semantic Space Models , 2007, CL.

[18]  Chih-Jen Lin,et al.  Projected Gradient Methods for Nonnegative Matrix Factorization , 2007, Neural Computation.

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

[20]  Pieter Adriaans,et al.  Qualia structures and their impact on the concrete noun categorization task , 2008 .

[21]  Jason Weston,et al.  A unified architecture for natural language processing: deep neural networks with multitask learning , 2008, ICML '08.

[22]  Marco Baroni,et al.  BagPack: A General Framework to Represent Semantic Relations , 2009, ArXiv.

[23]  Hinrich Schütze,et al.  Unsupervised Classification with Dependency Based Word Spaces , 2009 .

[24]  Eneko Agirre,et al.  A Study on Similarity and Relatedness Using Distributional and WordNet-based Approaches , 2009, NAACL.

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

[26]  Alessandro Lenci,et al.  Distributional Memory: A General Framework for Corpus-Based Semantics , 2010, CL.

[27]  Massimo Poesio,et al.  Strudel: A distributional semantic model based on properties and types , 2010 .

[28]  Yoshua Bengio,et al.  Word Representations: A Simple and General Method for Semi-Supervised Learning , 2010, ACL.

[29]  Rada Mihalcea,et al.  Semantic Relatedness Using Salient Semantic Analysis , 2011, AAAI.

[30]  Jason Weston,et al.  Natural Language Processing (Almost) from Scratch , 2011, J. Mach. Learn. Res..

[31]  Evgeniy Gabrilovich,et al.  Large-scale learning of word relatedness with constraints , 2012, KDD.

[32]  Katrin Erk,et al.  Vector Space Models of Word Meaning and Phrase Meaning: A Survey , 2012, Lang. Linguistics Compass.

[33]  Andrew Y. Ng,et al.  Semantic Compositionality through Recursive Matrix-Vector Spaces , 2012, EMNLP.

[34]  Andrew Y. Ng,et al.  Improving Word Representations via Global Context and Multiple Word Prototypes , 2012, ACL.

[35]  John A Bullinaria,et al.  Extracting semantic representations from word co-occurrence statistics: stop-lists, stemming, and SVD , 2012, Behavior research methods.

[36]  Mirella Lapata,et al.  A Comparison of Vector-based Representations for Semantic Composition , 2012, EMNLP.

[37]  Peter D. Turney.  Domain and Function: A Dual-Space Model of Semantic Relations and Compositions , 2012, J. Artif. Intell. Res..

[38]  Quoc V. Le,et al.  Exploiting Similarities among Languages for Machine Translation , 2013, ArXiv.

[39]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[40]  Marc'Aurelio Ranzato,et al.  DeViSE: A Deep Visual-Semantic Embedding Model , 2013, NIPS.

[41]  Andrew Y. Ng,et al.  Zero-Shot Learning Through Cross-Modal Transfer , 2013, NIPS.

[42]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[43]  Steven Skiena,et al.  The Expressive Power of Word Embeddings , 2013, ArXiv.

[44]  Geoffrey Zweig,et al.  Linguistic Regularities in Continuous Space Word Representations , 2013, NAACL.

[45]  Elia Bruni,et al.  Multimodal Distributional Semantics , 2014, J. Artif. Intell. Res..

[46]  Stephen Clark,et al.  Vector Space Models of Lexical Meaning , 2015 .