欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
分面导航利用动态多维分类目录组织查询结果，从而有效减轻数据库资源定位过程中的信息过载．现有的分面导航限制用户每次增删一个查询关键字，无法满足对具有丰富语义的导航操作的需求．另一方面，高效的动态目录生成算法的缺乏阻碍了分面导航在大规模数据中的应用．该文提出了层次概念格，对分面导航中不同浏览状态之间的关系进行建模．基于该层次概念格模型，该文设计了若干新的导航操作以支持用户在不同浏览状态之间更灵活地跳转，从而更有效地进行知识发现．为获取该层次概念格以支持导航的灵活性和实时性，该文提出了层次概念格的高效挖掘和索引算法L-Miner．L-Miner以深度优先方式挖掘所有节点，每得到一个新节点，就更新已挖掘节点之间的边．通过对底层格节点的倒排索引，L-Minder可以高效地进行边更新．实验结果表明：L—Miner的速度远快于现有算法，而其构建的索引结构的存储代价更低．
英文摘要：
Faceted navigation is effective in reducing source identification, by organizing query result into information overload in the process of redynamic multi-dimensional categories. Existing approaches allow users to add or delete only one query keyword at each step, which cannot meet the demands for semantic-rich navigation operations. Moreover, the lack of efficient algo- rithms for generating dynamic categories makes faceted navigation non-scalable to large datasets. This paper proposes a hierarchical concept lattice for modeling the relationship between different navigation states. Then, a series of navigation operations are proposed to support more flexible transitions between navigation states and hence achieve more effective knowledge discovery. To guarantee real-time response, this paper also devises an efficient algorithm L-Miner for mining and indexing the hierarchical concept lattice. L-Miner discovers all the nodes in a depth-first way and updates the edges between all the generated nodes each time a new node is detected. Empirical studies demonstrate that L-Miner is much faster than existing approaches, utilizing a much smaller indexing structure.
同期刊论文项目
同项目期刊论文
Quality-of-Service Analysis of Queuing Systems with Long-Range-Dependent Network Traffic and Variabl
Analytical Modelling and Optimization of Congestion Control for Prioritized Multi-Class Self-Similar
期刊信息
位置：