%0 Conference Paper %1 falk-lapesa-2022-scaling %A Falk, Neele %A Lapesa, Gabriella %B Proceedings of the Thirteenth Language Resources and Evaluation Conference %C Marseille, France %D 2022 %I European Language Resources Association %K argumentmining argumentquality deliberation %P 3301--3318 %T Scaling up Discourse Quality Annotation for Political Science %U https://aclanthology.org/2022.lrec-1.353 %X The empirical quantification of the quality of a contribution to a political discussion is at the heart of deliberative theory, the subdiscipline of political science which investigates decision-making in deliberative democracy. Existing annotation on deliberative quality is time-consuming and carried out by experts, typically resulting in small datasets which also suffer from strong class imbalance. Scaling up such annotations with automatic tools is desirable, but very challenging. We take up this challenge and explore different strategies to improve the prediction of deliberative quality dimensions (justification, common good, interactivity, respect) in a standard dataset. Our results show that simple data augmentation techniques successfully alleviate data imbalance. Classifiers based on linguistic features (textual complexity and sentiment/polarity) and classifiers integrating argument quality annotations (from the argument mining community in NLP) were consistently outperformed by transformer-based models, with or without data augmentation.

@inproceedings{falk-lapesa-2022-scaling, abstract = {The empirical quantification of the quality of a contribution to a political discussion is at the heart of deliberative theory, the subdiscipline of political science which investigates decision-making in deliberative democracy. Existing annotation on deliberative quality is time-consuming and carried out by experts, typically resulting in small datasets which also suffer from strong class imbalance. Scaling up such annotations with automatic tools is desirable, but very challenging. We take up this challenge and explore different strategies to improve the prediction of deliberative quality dimensions (justification, common good, interactivity, respect) in a standard dataset. Our results show that simple data augmentation techniques successfully alleviate data imbalance. Classifiers based on linguistic features (textual complexity and sentiment/polarity) and classifiers integrating argument quality annotations (from the argument mining community in NLP) were consistently outperformed by transformer-based models, with or without data augmentation.}, added-at = {2023-05-16T12:53:27.000+0200}, address = {Marseille, France}, author = {Falk, Neele and Lapesa, Gabriella}, biburl = {https://puma.ub.uni-stuttgart.de/bibtex/28e69c3c17b012edb190f0f72e683fcdc/neelefalk}, booktitle = {Proceedings of the Thirteenth Language Resources and Evaluation Conference}, interhash = {f9e299a7a1ea7c43636098a5d2a249d3}, intrahash = {8e69c3c17b012edb190f0f72e683fcdc}, keywords = {argumentmining argumentquality deliberation}, month = jun, pages = {3301--3318}, publisher = {European Language Resources Association}, timestamp = {2023-05-16T12:53:27.000+0200}, title = {Scaling up Discourse Quality Annotation for Political Science}, url = {https://aclanthology.org/2022.lrec-1.353}, year = 2022 }