Modeling of Knowledge with Imprecision in Linked Data Environment

网络变成了一个巨大的数据和信息储存库。它的不断增长给用户带来了很多机会，也带来了挑战：如何拥抱海量数据？如何找到新的信息？如何处理不精确和缺失的数据？如何将数据转化为信息，进而转化为知识？幸运的是，我们正处于数据在网络上表示和存储方式的重大和深远变化的开端。语义网的一个最基本的方面--资源描述框架--引起了人们的极大关注。资源描述框架的应用导致了一个真正的分布式和高度互联的数据网络。这创造了一个适合解决上述问题的环境。-- 拟议的研究项目特别强调建立、更新和利用基于网上获得的数据和信息建立的知识模型的过程。该项目的一个关键创新是将计算智能技术与网络技术相融合，以充分探索数据，包括时态数据，并利用资源描述框架的内在互联性。这些活动将导致建立连贯的知识创造过程和系统的雏形。简而言之，提出的方法侧重于使用1)资源描述框架数据的分层聚类来形成知识模型；2)构建聚类的自动概括；使用聚合和数据同化技术来维护模型的增量更新，该技术考虑了不同数据片段中的不精确度和置信度以及关于数据的时态信息；以及对知识模型的可视化和基于查询的探索，从而导致对数据和知识的多方面处理。 拟议的项目对目前和未来在智能网络系统领域的研究产生了直接和重要的影响。预计该项目将对侧重于建立新一代系统的方法作出重大贡献，这些系统支持用户从网上收集数据并对其进行处理以创造知识的活动。这可能导致建立个人知识库，帮助用户开展利用网上信息的活动。-- 该项目不仅将以创造知识产权(新技术、专利)的形式产生立竿见影的效果，而且还将为具备高度竞争力的技能的下一代工程师和研究人员做好准备。该项目的这些方面将支持加拿大成为顶尖技术先进国家之一的追求。该项目将有助于创造就业机会和发展先进技术。

The web becomes an enormous repository of data and information. Its constant growth creates a lot of opportunities as well as challenges for the users: How to embrace large amounts of data? How to find new pieces of information? How to deal with imprecision and missing data? How to convert data into information and then into knowledge? Fortunately, we are at the onset of significant and far-reaching changes in the way data are represented and stored on the web. One of the most fundamental aspects of the Semantic Web – Resource Description Framework – is generating a lot of attention. The application of Resource Description Framework induces a truly distributed and highly interconnected network of data. This creates an environment suitable for addressing the questions stated above.   The proposed research project puts a special emphasis on processes of constructing, updating and utilizing knowledge models built based on data and information obtained on the web. A key innovation of this project is a fusion of Computational Intelligence techniques with web technologies to fully explore data, including temporal data, and to take advantage of Resource Description Framework’s intrinsic interconnectivity. These activities will lead to establishing coherent rudiments of knowledge creation processes and systems. In a nutshell, the proposed methodology focuses on forming knowledge models using 1) a hierarchical clustering of Resource Description Framework data; 2) an automatic generalization of the constructed clusters; maintaining the models with incremental updates using aggregation and data assimilation techniques that take into account imprecision and confidence levels in different pieces of data, as well as temporal information about data; and visual and query-based exploration of knowledge models leading to multi-facet processing of data and knowledge.  The proposed project exhibits a direct and essential impact on the current and future research in the area of intelligent web systems. It is expected that the project will lead to significant contributions in methodologies focused on building new generation of systems that support the users in their activities related to collecting data from the web, and processing it towards creation of knowledge. This could lead to personal knowledge repositories that will help users with their activities of utilizing information available on the web.   The project would provide not only immediate results in the form of creation of intellectual property (new technologies, patents), but also in preparation of future generation of engineers and researchers equipped with highly competitive skills. These aspects of the project would support Canada’s pursuit to become one of the top technologically advanced countries. The project will contribute to the generation of employment and development of advanced technologies.