• DocumentCode
    3316894
  • Title

    Sentence semantic orientation calculation algorithm based on structure analysis

  • Author

    Yun Sha ; Jing Liu

  • Author_Institution
    Sch. of Inf. Eng., Beijing Inst. of Petrochem. Technol., Beijing, China
  • fYear
    2013
  • fDate
    23-24 Dec. 2013
  • Firstpage
    110
  • Lastpage
    113
  • Abstract
    It is urgent to find an automatic method of an automatic orientation calculation on the large-scale text in Chinese. Rather than comment their orientation directly in comment texts, people may express their orientation on topic by describe its features using orientation words. The describe word on features indicated their orientation. The orientation of comments on object is called topic orientation. Although the word orientation, also called word polarity, is the foundation for the text orientation analysis, the sentence structure also should be considered in this process. In this paper, text topic orientation is calculated by not only word orientation, but also sentence structure, and an Emotion Submission algorithm is proposed to calculate the sentence topic orientation. This method is suitable for goods comments. The experiment results show that the proposed algorithm is reasonable.
  • Keywords
    natural language processing; text analysis; Chinese language; automatic orientation calculation; comment text; emotion submission algorithm; large-scale text; sentence semantic orientation calculation algorithm; sentence structure analysis; sentence topic orientation; text orientation analysis; text topic orientation; word orientation; word polarity; Algorithm design and analysis; Classification algorithms; Dictionaries; Internet; Ontologies; Semantics; Speech; Sentence orientation; sentence structure; topic orientation analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Instrumentation and Measurement, Sensor Network and Automation (IMSNA), 2013 2nd International Symposium on
  • Conference_Location
    Toronto, ON
  • Type

    conf

  • DOI
    10.1109/IMSNA.2013.6742828
  • Filename
    6742828