Silvio Cordeiro

Predicting the Compositionality of Nominal Compounds: Giving Word Embeddings a Hard Time

Distributional semantic models (DSMs) are often evaluated on artificial similarity datasets containing single words or fully compositional phrases. We present a large-scale multilingual evaluation of DSMs for predicting the degree of semantic compositionality of nominal compounds on 4 datasets for English and French. We build a total of 816 DSMs and perform 2,856 evaluations using word2vec, GloVe, and PPMI-based models. In addition to the DSMs, we compare the impact of different parameters, such as level of corpus preprocessing, context window size and number of dimensions. The results obtained have a high correlation with human judgments, being comparable to or outperforming the state of the art for some datasets (Spearmans ρ=.82 for the Reddy dataset).

Joining Forces for Multiword Expression Identification

Multiword Expressions (MWEs) display some kind of linguistic and statistical markedness that may influence the effectiveness of techniques that automatically identify them in texts. While parsing-based techniques for MWE identification are considered to be better at handling long-distance dependencies, passivization and internal modification, statistics-based techniques use association measures to detect statistical markedness regardless of syntactic form. In this paper we compare these two approaches focusing on nominal compounds in Portuguese. We compare the accuracy of each method and propose that combining the strengths of both for increased accuracy.

UFRGS&LIF at SemEval-2016 Task 10: Rule-Based MWE Identification and Predominant-Supersense Tagging

This paper presents our approach towards the SemEval-2016 Task 10 - Detecting Minimal Semantic Units and their Meanings. Systems are expected to provide a representation of lexical semantics by (1) segmenting tokens into words and multiword units and (2) providing a supersense tag for segments that function as nouns or verbs. Our pipeline rule-based system uses no external resources and was implemented using the mwetoolkit. First, we extract and filter known MWEs from the training corpus. Second, we group input tokens of the test corpus based on this lexicon, with special treatment for non-contiguous expressions. Third, we use an MWE-aware predominant-sense heuristic for supersense tagging. We obtain an F-score of 51.48% for MWE identification and 49.98% for supersense tagging.

Filtering and Measuring the Intrinsic Quality of Human Compositionality Judgments

This paper analyzes datasets with numerical scores that quantify the semantic compositionality of MWEs. We present the results of our analysis of crowdsourced compositionality judgments for noun compounds in three languages. Our goals are to look at the characteristics of the annotations in different languages; to examine intrinsic quality measures for such data; and to measure the impact of filters proposed in the literature on these measures. The cross-lingual results suggest that greater agreement is found for the extremes in the compositionality scale, and that outlier annotation removal is more effective than outlier annotator removal.