DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Hyun, Soon-Joo | - |
dc.contributor.advisor | 현순주 | - |
dc.contributor.author | Lee, Jae-Ho | - |
dc.contributor.author | 이재호 | - |
dc.date.accessioned | 2011-12-30 | - |
dc.date.available | 2011-12-30 | - |
dc.date.issued | 2005 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=392561&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/55404 | - |
dc.description | 학위논문(석사) - 한국정보통신대학교 : 공학부, 2005, [ ix, 69 p. ] | - |
dc.description.abstract | Ontology-based context models are widely used in ubiquitous computing because they have advantages in the acquisition of conceptual context through inferencing, context sharing, and context reusing. Among the benefits, inferencing enables context-aware applications to use conceptual contexts which cannot be acquired by sensors. However, inferencing causes processing delay and thus becomes the major obstacle to the implementation of context-aware applications. The delay becomes longer as the amount of context data increases. In this thesis, we propose a context pre-fetching method to reduce the size of context data to be processed in a working memory in attempt to speed up the inferencing. We build query-trees to identify context data relevant to the queries of an application. The query-tree is a powerful static analyzing tool for encoding all derivations of a given set of queries. Thus, we can select ground context data used in deriving answers to the application’s queries before runtime by using the query-tree. However, in the case of queries in the context-aware application, there are many queries about context data acquired from the sensors. The context data from the sensors do not need to be placed in a working memory until the value of the context are taken from the sensors as they are meaningless while their real values are sensed. Accordingly, the context data acquired at runtime can be excluded from a working memory for the computation in static time. To filter them out, we apply the context types into the query-tree in query-tree build time. We have classified the context data into three type categories: sensed context, defined context, and deduced context. They make it possible to indicate when and how context can get a meaningful value. By introducing the context types into a query-tree, the proposed scheme filters out the context data in a query-tree which are meaningless before runtime. Maintaining the pre-fetched context data optimal in a working ... | eng |
dc.language | eng | - |
dc.publisher | 한국정보통신대학교 | - |
dc.subject | Ontology-based context management | - |
dc.subject | Enhancing inference performance | - |
dc.subject | 추론 성능 향상 | - |
dc.subject | 온톨로지 기반의 상황정보 관리 | - |
dc.title | Application-oriented context pre-fetch method for enhancing inference performance in ontology-based context management | - |
dc.title.alternative | 온톨로지 상황정보에서의 추론 속도 향상을 위한 어플리케이션 질의 기반의 상황정보 선인출 기법 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 392561/225023 | - |
dc.description.department | 한국정보통신대학교 : 공학부, | - |
dc.identifier.uid | 020034560 | - |
dc.contributor.localauthor | Hyun, Soon-Joo | - |
dc.contributor.localauthor | 현순주 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.