• DocumentCode
    3588869
  • Title

    Specification Model of Paragraph Summarization by Verbal Relationships: Objective, Cause, Consequence, Concurrence

  • Author

    Trung Tran ; Dang Tuan Nguyen

  • Author_Institution
    Fac. of Comput. Sci., Vietnam Nat. Univ., Ho Chi Minh City, Vietnam
  • fYear
    2014
  • Firstpage
    205
  • Lastpage
    210
  • Abstract
    The objective of this paper is to present a new model to summarize the sentences of the natural language paragraph. To perform, we construct a specification model, which is called Verbal Relationship Based Computational Model (VRBCM), based on specifying four inter-sentential relationships between each pair of consecutive sentences: objective, cause, consequence, concurrence. The main components of VRBCM model are: (i) a set of information representations of lexicons in the original paragraph, (ii) a set of inner relationships in each sentence in the original paragraph, (iii) a set of inter-sentential relationships between each pair of consecutive sentences in the original paragraph, (iv) a set of syntactic structures of sentences in the new paragraph. With the constructing method, VRBCM model can be applied for many different natural languages.
  • Keywords
    natural language processing; VRBCM model; cause sentence; concurrence sentence; consecutive sentence; consecutive sentences; consequence sentence; inner relationships; intersentential relationships; lexicon representation; natural language paragraph; objective sentence; paragraph summarization specification model; sentence summarization; syntactic sentence structures; verbal relationship-based computational model; Cities and towns; Computational modeling; Context; Finite element analysis; Information representation; Natural languages; Syntactics; discourse representation; meaning summarization; sentence generation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Artificial Intelligence, Modelling and Simulation (AIMS), 2014 2nd International Conference on
  • Print_ISBN
    978-1-4799-7599-0
  • Type

    conf

  • DOI
    10.1109/AIMS.2014.13
  • Filename
    7102461