一般注記type:text
Knowledge-collection and accumulation are essential functions for realizing a knowledge-creation environment. Our system for an objective-video with a story on the Internet focuses on expert-knowledge collection and accumulation based video contents with multimedia, such as images, texts and audio data. The purpose of this system design is to generate "relationships of uploaded data" across heterogeneous categories in order to associate among uploaded contents in various categories related to specific topics included in objective-video, in a cross-disciplinary way. Our system creates two types of "relationships of uploaded data". One type is automatically generated according to similarities among the contents of uploaded data. The other relationship is created by experts once they are shown a list of uploaded content in other categories in response to a query. An important feature of this system is a data structure for expressing uploaded multimedia data, as knowledge-collection related to an objective-video and is called a "multi-dimensional knowledge model" (MDKM). Our proposed MDKM can determine the similarity or association between two sets of uploaded data by computing the relevance score between uploaded data in various dimensions that are their corresponding metadata, such as tag, color histograms, and harmony. By using this system, experts upload knowledge in categories related to the objective-video and accumulate their knowledge. Our system is applicable to E-learning systems, Internet video-sharing platforms, and participatory entertainment systems.
2014年度慶應義塾大学JR東日本寄附講座報告書
慶應義塾大学交通運輸情報プロジェクト
その3 : 慶應義塾SFC大学院生•学部生の研究
identifier:SFC-RM2014-004
identifier:請求記号:
連携機関・データベース国立情報学研究所 : 学術機関リポジトリデータベース（IRDB）（機関リポジトリ）