學術產出-NSC Projects

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

  • No doi shows Citation Infomation
題名 應用Bayesian架構發展分類學習模型
其他題名 Categorization by local relationship among exemplars
作者 楊立行;許清芳;鄭惟尹;蔡涵如
Yang,Lee-Xieng
貢獻者 國立政治大學心智、大腦與學習研究中心
日期 2009
上傳時間 9-Oct-2013 17:27:31 (UTC+8)
摘要 The principal purpose of this study is to develop a categorization model with Bayesian
     theorems. The developed model SOE holds the assumptions that the percept of each
     input is represented as a normal distribution over the units on the stimulus dimension and
     that the classi cation is made by referencing the weighted strengths over the units within
     the scope of attention focus. SOE in this study is evident to be sensitive to the repetition
     of stimuli, stimulus order, and, the most important, the variability e ect in categorization.
     This e ect is thought to be a severe challenge to the exemplar-based models. However, the
     success of SOE suggests a possible exemplar-based account for this e ect.
關聯 執行期間:9708-9809
資料類型 report
dc.contributor 國立政治大學心智、大腦與學習研究中心en_US
dc.creator (作者) 楊立行;許清芳;鄭惟尹;蔡涵如zh_TW
dc.creator (作者) Yang,Lee-Xieng-
dc.date (日期) 2009en_US
dc.date.accessioned 9-Oct-2013 17:27:31 (UTC+8)-
dc.date.available 9-Oct-2013 17:27:31 (UTC+8)-
dc.date.issued (上傳時間) 9-Oct-2013 17:27:31 (UTC+8)-
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/61285-
dc.description.abstract (摘要) The principal purpose of this study is to develop a categorization model with Bayesian
     theorems. The developed model SOE holds the assumptions that the percept of each
     input is represented as a normal distribution over the units on the stimulus dimension and
     that the classi cation is made by referencing the weighted strengths over the units within
     the scope of attention focus. SOE in this study is evident to be sensitive to the repetition
     of stimuli, stimulus order, and, the most important, the variability e ect in categorization.
     This e ect is thought to be a severe challenge to the exemplar-based models. However, the
     success of SOE suggests a possible exemplar-based account for this e ect.
-
dc.format.extent 333103 bytes-
dc.format.mimetype application/pdf-
dc.language.iso en_US-
dc.relation (關聯) 執行期間:9708-9809en_US
dc.title (題名) 應用Bayesian架構發展分類學習模型zh_TW
dc.title.alternative (其他題名) Categorization by local relationship among exemplars-
dc.type (資料類型) reporten