Graph Ensemble Learning over Multiple Dependency Trees for Aspect-level Sentiment Classification
21649Authors
nan
Abstract
AbstractRecent work on aspect-level sentiment classification has demonstrated the efficacy of incorporating syntactic structures such as dependency trees with graph neural networks (GNN), but these approaches are usually vulnerable to parsing errors. To better leverage syntactic information in the face of unavoidable errors, we propose a simple yet effective graph ensemble technique, GraphMerge, to make use of the predictions from different parsers. Instead of assigning one set of model parameters to each dependency tree, we first combine the dependency relations from different parses before applying GNNs over the resulting graph. This allows GNN models to be robust to parse errors at no additional computational cost, and helps avoid overparameterization and overfitting from GNN layer stacking by introducing more connectivity into the ensemble graph. Our experiments on the SemEval 2014 Task 4 and ACL 14 Twitter datasets show that our GraphMerge model not only outperforms models with single dependency tree, but also beats other ensemble models without adding model parameters.
Keywords
Full Content
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 226–241 June 6–11, 2021. ©2021 Association for Computational Linguistics 226 Paragraph-level Rationale Extraction through Regularization: A case study on European Court of Human Rights Cases Ilias Chalkidis † ‡ Manos Fergadiotis † ‡ Dimitrios Tsarapatsanis ⋆ Nikolaos Aletras ⋄ Ion Androutsopoulos ‡ † Prodromos Malakasiotis † ‡ † EY AI Centre of Excellence in Document Intelligence, NCSR “Demokritos” ‡ Department of Informatics, Athens University of Economics and Business ⋄ Computer Science Department, University of Sheffield ⋆ Law School, University of York Abstract Interpretability or explainability is an emerg- ing research field in NLP. From a user-centric point of view, the goal is to build models that provide proper justification for their deci- sions, similar to those of humans, by requir- ing the models to satisfy additional constraints. To this end, we introduce a new application on legal text where, contrary to mainstream literature targeting word-level rationales, we conceive rationales as selected paragraphs in multi-paragraph structured court cases. We also release a new dataset comprising Euro- pean Court of Human Rights cases, includ- ing annotations for paragraph-level rationales. We use this dataset to study the effect of al- ready proposed rationale constraints, i.e., spar- sity, continuity, and comprehensiveness, for- mulated as regularizers. Our findings indicate that some of these constraints are not bene- ficial in paragraph-level rationale extraction, while others need re-formulation to better han- dle the multi-label nature of the task we con- sider. We also introduce a new constraint, sin- gularity, which further improves the quality of rationales, even compared with noisy rationale supervision. Experimental results indicate that the newly introduced task is very challenging and there is a large scope for further research. 1 Introduction Model interpretability (or explainability) is an emerging field of research in NLP (Lipton, 2018; Jacovi and Goldberg, 2020). From a model-centric point of view, the main focus is to demystify a model’s inner workings, for example targeting self- attention mechanisms (Jain and Wallace, 2019; Wiegreffe and Pinter, 2019), and more recently Transformer-based language models (Clark et al., 2019; Kovaleva et al., 2019; Rogers et al., 2020). From a user-centric point of view, the main fo- cus is to build models that learn to provide proper Correspondence to: ihalk.aueb.gr justification for their decisions, similar to those of humans, (Zaidan et al., 2007; Lei et al., 2016; Chang et al., 2019; Yu et al., 2019) by requiring the models to satisfy additional constraints. Here we follow a user-centric approach to ratio- nale extraction, where the model learns to select a subset of the input that justifies its decision. To this end, we introduce a new application on le- gal text where, contrary to mainstream literature targeting word-level rationales, we conceive ratio- nales as automatically selected paragraphs in multi- paragraph structured court cases. While previous related work targets mostly binary text classifica- tion tasks (DeYoung et al., 2020), our task is a highly skewed multi-label text …
Content truncated to 500 words
Labeled Text
AbstractRecent work on aspect-level sentiment classification has demonstrated the efficacy of incorporating syntactic structures such as dependency trees with graph neural networks (GNN), but these approaches are usually vulnerable to parsing errors. To better leverage syntactic information in the face of unavoidable errors, we propose a simple yet effective graph ensemble technique, GraphMerge, to make use of the predictions from different parsers. Instead of assigning one set of model parameters to each dependency tree, we first combine the dependency relations from different parses before applying GNNs over the resulting graph. This allows GNN models to be robust to parse errors at no additional computational cost, and helps avoid overparameterization and overfitting from GNN layer stacking by introducing more connectivity into the ensemble graph. Our experiments on the SemEval 2014 Task 4 and ACL 14 Twitter datasets show that our GraphMerge model not only outperforms models with single dependency tree, but also beats other ensemble models without adding model parameters.
Publication Details
- Paper ID:
- 21649
- Year:
- 2021
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan