• DocumentCode
    466968
  • Title

    Arithmetic Computing Based Chinese Automatic Parsing Method

  • Author

    Yang, Xiao ; Wan, Jiancheng ; Zhang, Ling

  • Author_Institution
    Shandong Univ., Jinan
  • Volume
    2
  • fYear
    2007
  • fDate
    July 30 2007-Aug. 1 2007
  • Firstpage
    242
  • Lastpage
    247
  • Abstract
    Cohesion degree is an important feature in differentiating ambiguous Chinese phrases. At the present time, there are no module in computer-oriented linguistic theories to represent this information and it is represented by additional statistical modules. The Binary Combinatorial Grammar (BCG) is based on arithmetic computing and integrates the cohesion degree as the priorities between operators. In order to resolve the share space problem and to improve the efficiency of operator-precedence shift-reduce parsing algorithm, the notion of precedence is introduced to the CYK algorithm to determine the combinatorial direction and to prune a portion of unnecessary subtrees during the parsing process. In the experiment of Chinese BCG parsing, all of the spent time, produced edges and result trees in the precedence based CYK algorithm are much less than that of the traditional CYK algorithm and the operator precedence based shift-reduce algorithm.
  • Keywords
    grammars; natural languages; arithmetic computing; binary combinatorial grammar; chinese automatic parsing method; cohesion degree; computer oriented linguistic theories; operator-precedence shift-reduce parsing algorithm; share space problem; statistical modules; Artificial intelligence; Computer networks; Computer science; Concurrent computing; Digital arithmetic; Distributed computing; Mathematics; Natural languages; Software engineering; Technical Activities Guide -TAG;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing, 2007. SNPD 2007. Eighth ACIS International Conference on
  • Conference_Location
    Qingdao
  • Print_ISBN
    978-0-7695-2909-7
  • Type

    conf

  • DOI
    10.1109/SNPD.2007.369
  • Filename
    4287686