Variants of Vector Space Reductions for Predicting the Compositionality of English Noun Compounds

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 4379–4387 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 4379 Variants of Vector Space Reductions for Predicting the Compositionality of English Noun Compounds Pegah Alipoor1, Sabine Schulte im Walde2 1Sharif University of Technology, Islamic Republic of Iran 2Institute for Natural Language Processing, University of Stuttgart, Germany 1palipoor@ce.sharif.edu 2schulte@ims.uni-stuttgart.de Abstract Predicting the degree of compositionality of noun compounds such as snowball and butterfly is a crucial ingredient for lexicography and Natural Language Processing applications, to know whether the compound should be treated as a whole, or through its constituents, and what it means. Computational approaches for an automatic prediction typically represent and compare compounds and their constituents within a vector space and use distributional similarity as a proxy to predict the semantic relatedness between the compounds and their constituents as the compound’s degree of compositionality. This paper provides a systematic evaluation of vector-space reduction variants across kinds, exploring reductions based on part-of-speech next to and also in combination with Principal Components Analysis using Singular Value Decomposition, and word2vec embeddings. We show that word2vec and nouns-only dimensionality reductions are the most successful and stable vector space reduction variants for our task. Keywords: Noun Compounds, Compositionality, Vector Spaces, Dimensionality Reduction 1. Introduction Predicting the degree of compositionality of noun com- pounds (and multi-word expressions in more general) is a crucial ingredient for lexicography and Natural Language Processing (NLP) applications, to know whether the ex- pression should be treated as a whole, or through its con- stituents, and what the expression means. Compare, for example, the English noun compounds snowball –a ball consisting of snow, where clearly both constituents snow and ball contribute to the meaning of the compound– and butterfly –where the semantic contribution of the modifier noun butter is not obvious without knowing about the et- ymology of the compound. Studies such as Cholakov and Kordoni (2014), Weller et al. (2014), Cap et al. (2015), and Salehi et al. (2015b) are examples of NLP applications that have integrated the prediction of multi-word composition- ality into statistical machine translation. Accordingly, the field has witnessed a rich amount of com- putational approaches to automatically predict the degree of compositionality of noun compounds. These approaches typically represent compounds and their constituents within a vector space, and then compare the compound vectors with the constituent vectors as a proxy to the compounds’ degree of compositionality (Reddy et al., 2011b; Reddy et al., 2011a; Salehi and Cook, 2013; Schulte im Walde et al., 2013; Salehi et al., 2014; Schulte im Walde et al., 2016; Cordeiro et al., 2019). Most of the approaches focus on English and German; most recently, Cordeiro et al. (2019) applied their framework to also French and Portuguese. All of the above-mentioned approaches explored variants of vector space models in some way, regarding the compos- ite functions to combine the constituent vectors (Reddy et al., 2011b); or regarding the translations of compounds and constituents into multiple …

Content truncated to 500 words