Exploring Vector Space Models to Predict the Compositionality of German Noun-Noun Compounds

This paper explores two hypotheses regarding vector space models that predict the compositionality of German noun-noun compounds: (1) Against our intuition, we demonstrate that window-based rather than syntax-based distributional features perform better predictions, and that not adjectives or verbs but nouns represent the most salient part-of-speech. Our overall best result is state-of-the-art, reaching Spearman’s = 0.65 with a wordspace model of nominal features from a 20word window of a 1.5 billion word web corpus. (2) While there are no significant differences in predicting compound‐modifier vs. compound‐head ratings on compositionality, we show that the modifier (rather than the head) properties predominantly influence the degree of compositionality of the compound.

[1]  Ulrich Heid,et al.  Collocations of Complex Nouns : Evidence for Lexicalisation , 2004 .

[2]  Sabine Schulte im Walde,et al.  The (Un)expected Effects of Applying Standard Cleansing Models to Human Ratings on Compositionality , 2013, MWE@NAACL-HLT.

[3]  Akiko Nagano,et al.  The Oxford Handbook of Compounding , 2010 .

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

[5]  Michael Schiehlen.  A Cascaded Finite-State Parser for German , 2003, EACL.

[6]  Marco Baroni,et al.  Nouns are Vectors, Adjectives are Matrices: Representing Adjective-Noun Constructions in Semantic Space , 2010, EMNLP.

[7]  Suresh Manandhar,et al.  An Empirical Study on Compositionality in Compound Nouns , 2011, IJCNLP.

[8]  Rochelle Lieber,et al.  The Oxford handbook of compounding , 2011 .

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

[10]  Suresh Manandhar,et al.  Dynamic and Static Prototype Vectors for Semantic Composition , 2011, IJCNLP.

[11]  Mirella Lapata,et al.  Composition in Distributional Models of Semantics , 2010, Cogn. Sci..

[12]  Silvia Bernardini,et al.  The WaCky wide web: a collection of very large linguistically processed web-crawled corpora , 2009, Lang. Resour. Evaluation.

[13]  Christina L. Gagné,et al.  Constituent integration during the processing of compound words: Does it involve the use of relational structures? , 2009 .

[14]  John Carroll,et al.  Detecting a Continuum of Compositionality in Phrasal Verbs , 2003, ACL 2003.

[15]  H. Schütze,et al.  Dimensions of meaning , 1992, Supercomputing '92.

[16]  Ulrich Heid,et al.  Design and Application of a Gold Standard for Morphological Analysis: SMOR as an Example of Morphological Evaluation , 2010, LREC.

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

[18]  Sabine Schulte im Walde,et al.  Determining the Degree of Compositionality of German Particle Verbs by Clustering Approaches , 2010, KONVENS.

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

[20]  Louise McNally,et al.  Intensionality was only alleged: On adjective-noun composition in distributional semantics , 2013, IWCS.

[21]  A. Stuart,et al.  Non-Parametric Statistics for the Behavioral Sciences. , 1957 .

[22]  Suzanne Stevenson,et al.  Classifying Particle Semantics in English Verb-Particle Constructions , 2006 .

[23]  Colin J. Bannard,et al.  Learning about the meaning of verb-particle constructions from corpora , 2005, Comput. Speech Lang..

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

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

[26]  SabineSchulteim Walde,et al.  Exploring Features to Identify Semanti Nearest Neighbours: A Case Study on German Parti le Verbs , 2012 .

[27]  Bernice W. Polemis.  Nonparametric Statistics for the Behavioral Sciences , 1959 .

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

[29]  Rochelle Lieber,et al.  Introduction: Status and Definition of Compounding , 2011 .

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

[31]  Dominic Widdows,et al.  Semantic Vector Products: Some Initial Investigations , 2008 .

[32]  Wolfgang Fleischer,et al.  Wortbildung der deutschen Gegenwartssprache , 2012 .

[33]  Christina L. Gagné,et al.  Inferential Processing and Meta-Knowledge as the Bases for Property Inclusion in Combined Concepts. , 2011 .

[34]  Curt Burgess,et al.  Producing high-dimensional semantic spaces from lexical co-occurrence , 1996 .

[35]  Verena Klos,et al.  5 Komposition und Kompositionalität: Desiderate und Hypothesen , 2011 .