A Corpus and Model Integrating Multiword Expressions and Supersenses

By Nathan Schneider, Noah A. Smith. 2015

85 citations recorded by the source corpus.

This paper introduces a task of identifying and semantically classifying lexical expressions in running text. We investigate the online reviews genre, adding semantic supersense annotations to a 55,000 word English corpus that was previously annotated for multiword expressions. The noun and verb supersenses apply to full lexical expressions, whether single- or multiword. We then present a sequence tagging model that jointly infers lexical expressions and their supersenses. Results show that even with our relatively small training corpus in a noisy domain, the joint task can be performed to attain 70% class labeling F1.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems (2019, methodology): “6Tasks like QuAC (Choi et al., 2018a) and STREUSLE (Schneider and Smith, 2015) differed substantially from the format of other tasks in SuperGLUE, which we worried would incentivize users to spend significant effort on task-specific model designs, rather than focusing on general-purpose techniques.”
Identifying beneficial task relations for multi-task learning in deep neural networks (2017, methodology): “MWE Detection (MWE) We use the Streusle corpus (Schneider and Smith, 2015) to learn to identify multi-word expressions (on my own, cope with).”
Multi-Task Learning for Sequence Tagging: An Empirical Study (2018, methodology): “We perform multi-word expression identiﬁcation ( MWE ) and supersense tagging ( SUPSENSE ) on sentences from the reviews section of the English Web Treebank, annotated under the Streusle project (Schneider and Smith, 2015) 1 .”
SenseBERT: Driving Some Sense into BERT (2019, background): “Disambiguating a word’s supersense has been widely studied as a fundamental lexical cat-a egorization task [Ciaramita and Johnson, 2003, Basile, 2012, Schneider and A Smith, 2015].”
Supersense Embeddings: A Unified Model for Supersense Interpretation, Prediction, and Utilization (2016, methodology): “regained its popularity (Johannsen et al., 2014; Schneider and Smith, 2015), since supersenses provide disambiguating information, useful for numerous downstream NLP tasks, without the need of tedious fine-grained WSD.”
Multi-Task Learning of Keyphrase Boundary Classification (2017, methodology): “(2010), (4) identification of multi-word expressions using the Streusle corpus (Schneider and Smith, 2015); and (5) semantic super-sense tagging using the Semcor dataset, following Johannsen et al.”
Still a Pain in the Neck: Evaluating Text Representations on Lexical Composition (2019, background): “Phrase Type STREUSLE (Schneider and Smith, 2015) 3017/372/376 sentence s label per token O”
Low Resource Multi-Task Sequence Tagging - Revisiting Dynamic Conditional Random Fields (2020, background): “Streusle The Streusle dataset (Schneider and Smith, 2015) consists of three tasks.”