![]() ![]() We release the newly created corpus UKPConvArg1 and theĮxperimental software under open licenses. We experiment withįeature-rich SVM and bidirectional LSTM and obtain 0.76-0.78 accuracy and 0.35-0.40 Spearman'sĬorrelation in a cross-topic evaluation. (2) ranking all arguments to the topic based on their convincingness. We propose two tasks: (1) predicting which argument from an argument pair is more convincing and Ordering these findings are used as global constraints for cleaning the crowdsourced data. ![]() Investigate whether the relation "A is more convincing than B" exhibits properties of total We annotate a large datasets of 16k pairs of arguments over 32 topics and Problem as relation classification, where a pair of arguments having the same stance to the same Investigate qualitative properties of Web arguments, namely their convincingness. ![]() **Abstract:** We propose a new task in the field of computational argumentation in which we
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |