• DocumentCode
    3241769
  • Title

    Novel methods for energy-based cultural modeling and simulation: Why eight is great in Chinese culture

  • Author

    Olsher, Daniel J. ; Toh Heng Guan

  • Author_Institution
    Cognitive Sci. Program, Temasek Labs. Nat. Univ. of Singapore, Singapore, Singapore
  • fYear
    2013
  • fDate
    16-19 April 2013
  • Firstpage
    74
  • Lastpage
    81
  • Abstract
    A key frontier in artificial intelligence systems is the ability to represent and reason about culture, a notoriously difficult notion to define and operationalize. This paper presents a framework for representing, simulating, and reasoning on nuanced cultural worldviews, including culturally-mediated perception and judgment simulation, grounded in the INTELNET nuanced Energy-Based Knowledge Representation (EBKR) and COGVIEW conceptual-psychological formalisms. A detailed example cultural model is provided, covering aspects of Chinese culture relevant to lucky numbers. Simulation output includes information regarding emotions generated, involvement of key cultural notions, areas of agreement and disagreement, and other useful semantic attributes. Goals include the use of cultural worldview networks in reasoning and the simulation of judgments and emotions generated by stimuli in particular cultural contexts. Such frameworks can help simulate the cultural consequences of particular actions or ideas, determining when insults are likely to arise due to impingement on culturally important issues.
  • Keywords
    history; knowledge representation; COGVIEW conceptual-psychological formalisms; Chinese culture; INTELNET; artificial intelligence systems; cultural worldview networks; energy-based cultural modeling; energy-based knowledge representation; useful semantic attributes; Artificial intelligence; Cognition; Context; Context modeling; Cultural differences; Psychology; Semantics; Belief; Culture; Emotion Simulation; Energy-Based Knowledge Representation; Reasoning;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational Intelligence for Human-like Intelligence (CIHLI), 2013 IEEE Symposium on
  • Conference_Location
    Singapore
  • Type

    conf

  • DOI
    10.1109/CIHLI.2013.6613268
  • Filename
    6613268