Advanced search
1 file | 275.07 KB Add to list

LT3 at SemEval-2020 Task 7 : comparing feature-based and transformer-based approaches to detect funny headlines

Bram Vanroy (UGent) , Sofie Labat (UGent) , Olha Kaminska (UGent) , Els Lefever (UGent) and Veronique Hoste (UGent)
Author
Organization
Abstract
This paper presents two different systems for the SemEval shared task 7 on Assessing Humor in Edited News Headlines, sub-task 1, where the aim was to estimate the intensity of humor generated in edited headlines. Our first system is a feature-based machine learning system that combines different types of information (e.g. word embeddings, string similarity, part-of-speech tags, perplexity scores, named entity recognition) in a Nu Support Vector Regressor (NuSVR). The second system is a deep learning-based approach that uses the pre-trained language model RoBERTa to learn latent features in the news headlines that are useful to predict the funniness of each headline. The latter system was also our final submission to the competition and is ranked seventh among the 49 participating teams, with a root-mean-square error (RMSE) of 0.5253.
Keywords
natural language processing, lt3, sentiment analysis, transformers

Downloads

  • Vanroy et al - 2020 - LT3 at SemEval-2020 Task 7 - Comparing feature-based and transformer-based approaches to detect funny headlines.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 275.07 KB

Citation

Please use this url to cite or link to this publication:

MLA
Vanroy, Bram, et al. “LT3 at SemEval-2020 Task 7 : Comparing Feature-Based and Transformer-Based Approaches to Detect Funny Headlines.” Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020), International Committee for Computational Linguistics, 2020, pp. 1033–40.
APA
Vanroy, B., Labat, S., Kaminska, O., Lefever, E., & Hoste, V. (2020). LT3 at SemEval-2020 Task 7 : comparing feature-based and transformer-based approaches to detect funny headlines. Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020), 1033–1040. International Committee for Computational Linguistics.
Chicago author-date
Vanroy, Bram, Sofie Labat, Olha Kaminska, Els Lefever, and Veronique Hoste. 2020. “LT3 at SemEval-2020 Task 7 : Comparing Feature-Based and Transformer-Based Approaches to Detect Funny Headlines.” In Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020), 1033–40. International Committee for Computational Linguistics.
Chicago author-date (all authors)
Vanroy, Bram, Sofie Labat, Olha Kaminska, Els Lefever, and Veronique Hoste. 2020. “LT3 at SemEval-2020 Task 7 : Comparing Feature-Based and Transformer-Based Approaches to Detect Funny Headlines.” In Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020), 1033–1040. International Committee for Computational Linguistics.
Vancouver
1.
Vanroy B, Labat S, Kaminska O, Lefever E, Hoste V. LT3 at SemEval-2020 Task 7 : comparing feature-based and transformer-based approaches to detect funny headlines. In: Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020). International Committee for Computational Linguistics; 2020. p. 1033–40.
IEEE
[1]
B. Vanroy, S. Labat, O. Kaminska, E. Lefever, and V. Hoste, “LT3 at SemEval-2020 Task 7 : comparing feature-based and transformer-based approaches to detect funny headlines,” in Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020), Barcelona, Spain, 2020, pp. 1033–1040.
@inproceedings{8683420,
  abstract     = {{This paper presents two different systems for the SemEval shared task 7 on Assessing Humor in Edited News Headlines, sub-task 1, where the aim was to estimate the intensity of humor generated in edited headlines. Our first system is a feature-based machine learning system that combines different types of information (e.g. word embeddings, string similarity, part-of-speech tags, perplexity scores, named entity recognition) in a Nu Support Vector Regressor (NuSVR). The second system is a deep learning-based approach that uses the pre-trained language model RoBERTa to learn latent features in the news headlines that are useful to predict the funniness of each headline. The latter system was also our final submission to the competition and is ranked seventh among the 49 participating teams, with a root-mean-square error (RMSE) of 0.5253.}},
  author       = {{Vanroy, Bram and Labat, Sofie and Kaminska, Olha and Lefever, Els and Hoste, Veronique}},
  booktitle    = {{Proceedings of the 14th International Workshop on Semantic Evaluation (SemEval 2020)}},
  isbn         = {{9781952148316}},
  keywords     = {{natural language processing,lt3,sentiment analysis,transformers}},
  language     = {{eng}},
  location     = {{Barcelona, Spain}},
  pages        = {{1033--1040}},
  publisher    = {{International Committee for Computational Linguistics}},
  title        = {{LT3 at SemEval-2020 Task 7 : comparing feature-based and transformer-based approaches to detect funny headlines}},
  url          = {{https://www.aclweb.org/anthology/2020.semeval-1.0.pdf#page.1033}},
  year         = {{2020}},
}