We ran experiments on six combinations of features to predict sequences with Hierarchical Agglomerative Clustering. These features were transformed using tf-idf and word embedding feature selection was done using Principal Component Analysis (PCA). Syntactic features were extracted using dependency parsing, while dialogue control functions were manually labelled. This work proposes a framework to predict sequences in dialogues, using turn based syntactic features and dialogue control functions. Proceedings of the First Workshop on Computational Approaches to DiscourseĪssociation for Computational Linguistics An analysis of the clustering results indicate that using word embeddings and syntactic features, significantly improved the results.",īeyond Adjacency Pairs: Hierarchical Clustering of Long Sequences for Human-Machine Dialogues
Publisher = "Association for Computational Linguistics",Ībstract = "This work proposes a framework to predict sequences in dialogues, using turn based syntactic features and dialogue control functions.
Dialogue 1.2 mods#
| EMNLP SIG: Publisher: Association for Computational Linguistics Note: Pages: 11–19 Language: URL: DOI: 10.18653/v1/di-1.2 Bibkey: maitreyee-2020-beyond Copy Citation: BibTeX MODS XML Endnote More options… PDF: = "Beyond Adjacency Pairs: Hierarchical Clustering of Long Sequences for Human-Machine Dialogues",īooktitle = "Proceedings of the First Workshop on Computational Approaches to Discourse", Anthology ID: di-1.2 Volume: Proceedings of the First Workshop on Computational Approaches to Discourse Month: November Year: 2020 Address: Online Venues: CODI An analysis of the clustering results indicate that using word embeddings and syntactic features, significantly improved the results.
Abstract This work proposes a framework to predict sequences in dialogues, using turn based syntactic features and dialogue control functions.