Pavlina Synkova |  Jiri Mirovsky... -  《2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation,Vol. 2》 -  Joint International Conference on Computational Linguistics, Language Resources and Evaluation - 2024, - 1270～1279 - 共10页

摘要： trees), but also in the Penn Discourse Treebank 3.0 | We present the Prague Discourse Treebank 3.0 |  - a new version of the annotation of discourse |  relations marked by primary and secondary discourse |  connectives in the data of the Prague Dependency Treebank

关键词： discourse relations |  pragmatic relations |  Prague Discourse Treebank |  Penn Discourse Treebank

Jiri Mirovsky |  Pavlina Synkova... -  《2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation,Vol. 6》 -  Joint International Conference on Computational Linguistics, Language Resources and Evaluation - 2024, - 4067～4077 - 共11页

摘要： projection from the Penn Discourse Treebank 3.0, () manual |  plain texts) with the Penn Discourse Treebank 3.0 |  Street Journal part of the Penn Treebank. We use three |  of discourse types, and in the Penn format (on |  obtaining a high-quality annotation of explicit discourse

关键词： discourse relations |  annotation projection |  cost-effective annotation |  Prague Czech-English Dependency Treebank

WEI XIANG |  BANG WANG -  《ACM computing surveys》 - 2023,55(12) - 258.1～258.34 - 共34页 - 被引量：4

摘要：A discourse containing one or more sentences |  understanding of the theme of a discourse should take into |  implicit discourse relation recognition (IDRR) is to |  future research directions for discourse relation |  describes daily issues and events for people to

关键词： Implicit discourse relation |  relation recognition |  Penn discourse TreeBank |  natural language processing

Kezban Ba??büyük |  Deniz Zeyrek -  《Language resources and evaluation》 - 2023,57(1) - 223～256 - 共34页 - 被引量：1

摘要： annotated in the Penn Discourse TreeBank style. The paper |  the discourse usage of Turkish connectives, which |  disambiguate their discourse usage. The linguistic rules are |  the Turkish section of TED-Multilingual Discourse |  Bank and Turkish Discourse Bank 1.1, two datasets

关键词： Discourse processing |  Discourse connectives |  Usage disambiguation |  Connective lexicon |  Turkish

Jing Xu |  Ruifang He... -  《2023 International Joint Conference on Neural Networks: IJCNN 2023, Gold Coast, Australia, 18-23 June 2023, [v.1]》 -  International Joint Conference on Neural Networks - 2023, - 1～7 - 共7页

摘要： Discourse TreeBank (PDTB) demonstrate the effectiveness of | Implicit discourse relation recognition aims |  discourse arguments, which is a classification task | . Existing models mostly model the interaction of discourse |  patterns between two arguments under different discourse

关键词： Semantics |  Neural networks |  Routing |  Pattern recognition |  Iterative methods |  Task analysis

Lyu, Chen |  Feng, Wenhe -  《Neurocomputing》 - 2023,519(Jan.28) - 82～93 - 共12页

摘要： discourse treebank (PDTB). The main problem of the | Discourse structure is generally represented |  representations are rhetorical structure theory (RST) and Penn |  semantic relevance between the elementary discourse units |  EDUs. Discourse dependency structure (DDS) has been

关键词： Discourse structure |  Clause relevance structure |  Corpus construction |  Clause relevance recognition

Ma, Yuhao |  Zhu, Jian... -  《Applied Intelligence: The International Journal of Artificial Intelligence, Neural Networks, and Complex Problem-Solving Technologies》 - 2022,52(7) - 7700～7712 - 共13页 - 被引量：4

摘要： classification. Experimental results on Penn Discourse Treebank | Implicit discourse relation classification is |  one of the most challenging tasks in discourse | , classifying discourse relations usually requires |  classify multi-level discourse relations, improving the

关键词： Implicit discourse relation classification |  Bidirectional gated recurrent unit |  Graph attention network |  Hyperbolic spaces |  Discourse parsing

Amalia Mendes |  Pierre Lejeune -  《Computational Processing of the Portuguese Language: 15th International Conference on Computational Processing of the Portuguese Language (PROPOR 2022), March 21–23, 2022, Fortaleza, Brazil》 -  International Conference on Computational Processing of the Portuguese Language - 2022, - 79～89 - 共11页

摘要：). CRPC-DB follows the Penn Discourse Treebank style of | We present a new resource for discourse |  studies in Portuguese, the CRPC Discourse Bank (CRPC-DB |  and didactic/scientific texts. The discourse bank |  of 14,436 discourse relations. We present the main

关键词： Discourse bank |  Discourse relations |  Text coherence |  PDTB-style of annotation

NSTL主题词： Debates

Christian Chiarcos -  《Language Resources and Evaluation Conference, Vol. 4: 13th Language Resources and Evaluation Conference (LREC 2022), 20-25 June 2022, Marseille, France》 -  Language Resources and Evaluation Conference - 2022, - 2401～2412 - 共12页

摘要： relation inventory of the Penn Discourse Treebank (PDTB | Discourse marker inventories are lexical |  resources that define the meaning of discourse cues |  (discourse markers) in terms of associated discourse |  development of both discourse parsers and corpora with

关键词： discourse marker |  lexical knowledge graphs |  lexical induction |  OntoLex

Zujun Dou |  Yu Hong... -  《2022 IEEE 34th International Conference on Tools with Artificial Intelligence: ICTAI 2022, Macao, China, 31 October - 2 November 2022, [v.2]》 -  IEEE International Conference on Tools with Artificial Intelligence - 2022, - 500～507 - 共8页

摘要： on Penn Discourse TreeBank Corpus of version 2.0 | Implicit discourse relation classification | , decoding discourse relations heavily relies on the |  refers to a task of automatically determining |  relationships between arguments. It has been widely proven

关键词： Representation learning |  Recurrent neural networks |  Semantics |  Knowledge graphs |  Logic gates |  Encoding |  Decoding