An efficient query indexing mechanism for filtering geo-textual data
An efficient query indexing mechanism for filtering geo-textual data
复制标题
DOI：
10.1145/2463676.2465328
复制
发表时间：
2013-06
期刊：
影响因子：
--
通讯作者：
Lisi Chen;G. Cong;Xin Cao
中科院分区：
文献类型：
--
作者：
Lisi Chen;G. Cong;Xin Cao
地理标记和文本信息相关的大量数据正在以前所未有的规模生成。用户可能希望在一段时间内收到有趣的地理文本对象的通知。例如，用户可能希望在接下来的72小时内，当包含“garage sale”的tweet在距其住所5公里的范围内发布时，他能收到通知。在本文中，我们首次研究了一组传入的布尔范围连续查询流与一组传入的地理文本对象的实时匹配问题。我们开发了一套解决这个问题的新系统。特别是，我们提出了一种称为iq树的混合索引和新的成本模型，用于管理传入的布尔范围连续查询流。我们还提出了基于索引将查询与传入的地理文本对象匹配的算法。对所提出的技术实施的实证研究结果表明，本文的建议具有可扩展性和卓越的性能。
Massive amount of data that are geo-tagged and associated with text information are being generated at an unprecedented scale. Users may want to be notified of interesting geo-textual objects during a period of time. For example, a user may want to be informed when tweets containing term "garage sale" are posted within 5 km of the user's home in the next 72 hours. In this paper, for the first time we study the problem of matching a stream of incoming Boolean Range Continuous queries over a stream of incoming geo-textual objects in real time. We develop a new system for addressing the problem. In particular, we propose a hybrid index, called IQ-tree, and novel cost models for managing a stream of incoming Boolean Range Continuous queries. We also propose algorithms for matching the queries with incoming geo-textual objects based on the index. Results of empirical studies with implementations of the proposed techniques demonstrate that the paper's proposals offer scalability and are capable of excellent performance.