Semantic Expansion of Geographic Web Queries Based on Natural Language Positioning Expressions

The need for better Web search tools is getting increasing attention nowadays. About 20% of the queries currently submitted to search engines include geographic references. Thus, it is particularly important to work with the semantics of such queries, both by understanding the terminology and by recognizing geographic references in natural language text. In this paper, we explore the use of natural language expressions, which we call positioning expressions, to perform geographic searches on the Web, without resorting to geocoded data or gazetteers. Such positioning expressions denote the location of a subject of interest with respect to a landmark. Our approach leads to a query expansion technique that can be explored by virtually any keyword‐based search engine. Results obtained in our experiments show an expressive improvement over the traditional keyword‐based search and a potential path for tackling many kinds of common geographic queries.     

Supporting Spatial Semantics with SPARQL

Though the intersection of spatial data and semantic web technologies holds significant promise, there are still many challenges before this promise can be realized. One of these challenges is query representation. History suggests that an appropriate solution is a specialized query language for spatial data; however, with a broad interpretation of the SPARQL specification and extensions that would be useful outside the spatial realm, one can use SPARQL to query spatial concepts effectively. This article establishes a set of desiderata for a query language capable of dealing with spatial Semantic Web‐based data, discusses the challenges facing such a query language, and addresses these challenges with straightforward solutions that are broadly applicable. The effectiveness of these extensions is demonstrated using example queries.     

A parallel approach for improving Geo-SPARQL query performance

Geospatial Semantic Web promises better retrieval geospatial information for Digital Earth systems by explicitly representing the semantics of data through ontologies. It also promotes sharing and reuse of geospatial data by encoding it in Semantic Web languages, such as RDF, to form geospatial knowledge base. For many applications, rapid retrieval of spatial data from the knowledge base is critical. However, spatial data retrieval using the standard Semantic Web query language – Geo-SPARQL – can be very inefficient because the data in the knowledge base are no longer indexed to support efficient spatial queries. While recent research has been devoted to improving query performance on general knowledge base, it is still challenging to support efficient query of the spatial data with complex topological relationships. This research introduces a query strategy to improve the query performance of geospatial knowledge base by creating spatial indexing on-the-fly to prune the search space for spatial queries and by parallelizing the spatial join computations within the queries. We focus on improving the performance of Geo-SPARQL queries on knowledge bases encoded in RDF. Our initial experiments show that the proposed strategy can greatly reduce the runtime costs of Geo-SPARQL query through on-the-fly spatial indexing and parallel execution.     

GeoSQL: 一种可视化空间扩展SQL查询语言

根据结构化查询语言ＳＱＬ的基本原理,设计了空间查询语言ＧｅｏＳＱＬ及其表达形式、界面设计和实现方法。ＧｅｏＳＱＬ的特色为：①采用子查询（ｓｕｂｑｕｅｒｙ） 重组ＦＲＯＭ 语句,使其符合一般空间操作的程序;②ＧｅｏＳＱＬ的界面设计融入了Ｉｃｏｎ、Ｌｉｓｔｂｏｘ 等可视化部件,易于查询文本的形成,且减少了语法错误。     

空间信息自然语言查询接口的研究与应用

提出了空间信息自然语言查询接口(SINLQI)，并讨论了基于E—R语义词典的建立、中文分词、查询文法规则及其应用领域等主要问题。     

空间信息网格中远程空间连接查询的优化

随着空间信息网格的建设,网格平台上管理的空间信息资源越来越丰富,这促进了空间信息网格中空间数据分布式查询的应用需求,而在分布式空间查询中,空间连接查询操作往往成为性能的瓶颈.根据空间信息的特点,通过利用网格计算资源来优化空间连接查询的执行.首先基于网格服务构建网格平台分布式空间数据查询软件结构,通过设计远程空间连接执行服务利用网格平台中的计算资源;根据空间信息的特点.采用基于Kd-Tree空间分区并行连接的方法提高远程空间数据连接操作执行效率,并给出了远程空间连接执行的查询代价模型;然后根据连接代价模型设计了远程空间连接查询执行计划优化生成算法;最后总结了本文工作并探讨了下一步研究方向.     

A New GIScience Application for Visualized Natural Resources Management and Decision Support

This article presents two decision support plug‐ins in free GIS software: ArcGIS Explorer Desktop. The ubiquitous WebGIS Analysis Toolkit for Extensive Resources (uWATER) focuses on providing GIS analysis functions: spatial and attribute queries in the public services tool for users with limited access to commercial GIS software. The interface of uWATER is general enough to support decision‐making in numerous management issues in natural resources, economics and agriculture. The uWATER‐Pumping Assessment (uWATER‐PA) toolkit, on the other hand, is an extended package targeting the specific environmental issue of groundwater pumping impacts. The uWATER‐PA package is an excellent alternative to evaluating complex groundwater pumping assessment issues before investing significant time, labor, and funds in monitoring and detailed scientific study. It incorporates simulation of the physics of groundwater flow and user interaction into GIS software. A graphical user interface makes both data entry and interpretation of results intuitive to non‐technical individuals. Results are presented as colored drawdown maps and can be saved in GIS format for future dissemination. The impact of drawdown on existing wells can be characterized and mapped, through the use of uWATER's spatial query capabilities and the drawdown maps generated by uWATER‐PA.     

自然语言理解的中文地址匹配算法

地址匹配算法是位置服务的核心关键技术,具有广泛应用前景。在分析现有三类主要的中文地址匹配算法——要素层级匹配法、全文检索法、正则表达式法的基础上,本文提出了基于自然语言理解的中文地址匹配算法。在这个新算法中,建立了空间关系地址模型以解决中文地址抽象问题、地址库逻辑模型以解决地址信息的空间知识表达问题。新算法的完整流程包括 “预处理”、“地址解析”、“地址要素标准化”、“推理匹配”和“匹配登记”等五个环节,本文重点阐述了“地址解析”和“推理匹配”这两个重要环节,分别依据“自然语言理解”中的“中文分词”和“语义推理”原理,对用非结构化的中文自然语言来描述的中文地址进行处理,实现自然语言理解方法与地址匹配之间的结合,从而建立完整的基于自然语言理解的中文地址匹配算法。为验证该算法,开发了“中文地址智能匹配实验系统”,对河南省濮阳市人口库1000条居民地址数据进行匹配,匹配率达到了95%,准确率高于93%。     

Recent Literature

Development of a temporal geographic information system (GIS) and spatio-temporal data modeling are key to incorporating time into geographic information science. This paper describes how to design and develop temporal GIS that will work with spatio-temporal data represented in various data models, and it introduces a prototype temporal GIS with a case study. In temporal GIS, the integration of multiple spatio-temporal representations is based on common spatial and temporal reference systems. In other words, a map window of temporal GIS visualizes spatio-temporal data valid at the same time within one spatial area. To achieve such visualization, separate data editing and query modules are required for each spatio-temporal data model (STDM). In the temporal query interface, after a user specifies a time, the system automatically hires correspondent modules to retrieve spatio-temporal data valid at that time. Besides temporal queries common to all STDMs, each module may provide additional temporal query capabilities specific to that STDM. In the case study, I implement a prototype temporal GIS for three STDMs. The examples of query and visualization, which use three datasets (census data, land use/land cover, and elevation data) demonstrate the prototype temporal GIS can integrate multiple temporal representations.     

GQL: Extending Xquery to query GML documents

GML is becoming the de facto standard for electronic data exchange among the applications of Web and distributed geographic information systems. However, the conventional query languages (e.g. SQL and its extended versions) are not suitable for direct querying and updating of GML documents. Even the effective approaches working well with XML could not guarantee good results when applied to GML documents. Although XQuery is a powerful standard query language for XML, it is not proposed for queryings spatial features, which constitute the most important components in GML documents. We proposes GQL, a query language specification to support spatial queries over GML documents by extending XQuery. The data model, algebra, and formal semantics as well as various spatial functions and operations of GQL are presented in detail.     

自然空间查询语言解译机制研究

基于自然语言的空间查询语言是空间数据库和智能GIS的重要研究领域。针对空间查询语言的特殊性就基于自然语言的空间查询语言的解译机制进行了细致的研究。在空间目标名称库、空间词汇库、空间查询句型模板库、空间语料库、查询结果模板库等空间知识库的支持下,介绍了如何用最大匹配分词技术对空间查询语句进行分词;说明了利用空间查询句型模板库中的句型对查询语句进行句法分析,最终解译出空间查询目标和相应空间操作的过程;详细阐述了如何进一步将这些空间目标和空间操作转换成中间空间查询语句(即扩展空间数据类型和空间函数的SQL),并由关系数据库执行查询;最后就如何选择文字或图形两种语言形式进行查询结果表示做了说明。     

GQL： Extending XQuery to Query GML Documents

IntroductionThe geography markup language (GML)[1]is anXML encoding in compliance with ISO 19118 forthe transport and storage of geographic informationmodel according to the conceptual modeling frame-work used in the ISO 19100 series and includingboth the…     

受限自然语言查询在GIS中的应用

简单论述了空间信息系统的自然语言查询问题；在应用范围受限的条件下，提出了主要以自然语言中的关键词构造模式库来反映射查询语句的思想方法；建立了系统的体系结构，并对空间操作算子、空间查询语句、模式库的结构等关键问题进行了讨论。     

GIS空间数据研究

GIS是随着计算机技术发展而形成的一门新兴技术,其应用程度和范围也随之渗透延伸,正处于急剧变化与发展之中。阐述了GIS领域发展现状,介绍了GIS的进展,结合当前信息技术发展的特点。如空间数据误差或不确定性引起的拓扑关系不确定性,空间推理、空间查询处理、空间查询语言等与拓扑关系理论发展,GIS的发展趋势及其21世纪所面临的问题进行了分析和研究     

GIS中文查询系统中SQL语句的形成

由于GIS中文查询语句与SQL语句相差很大,直接转化非常困难,所以需要有某种中间逻辑形式语言。本文首先分析了查询语句的句型和基于此句型的栈式中间语言结构,然后重点研究了这种中间语言到SQL语句的转换。在此对SQL语句的结构中的查询条件段、查询目标段、分组段的判断策略进行了研究,难点在于嵌套结构的判断,在此借助中间关系进行转换,系统设计了相应的转换算法,并对算法进行了测试,测试表明系统所提出的转换策略和算法对大多数查询语句能转换为相应的SQL语句,为该课题的进一步研究奠定了基础。     

Indexing and querying moving objects with uncertain speed and direction in spatiotemporal databases

Efficient processing of spatiotemporal queries over moving objects with uncertainty has become imperative due to the increasing need for real-time information in highly dynamic environments. Most of the existing approaches focus on designing an index structure for managing moving objects with uncertainty and then utilize it to improve the query performance. All the proposed indexes, however, have their own limitations. In this paper, we devote to developing an efficient index, named the R ^lsd -tree, to index moving objects with uncertain speed and direction varying within respective known ranges. We design several pruning criteria combined with the R ^lsd -tree to answer the probabilistic range queries. Moreover, two models, the sampling-based probability model and the ER-based probability model, are proposed to quantify the possibility of each object being the query result. Finally, a thorough experimental evaluation is conducted to show the merits of the proposed techniques.     

Ontology‐supported Querying of Geographical Databases

Querying geographical information systems has been recognized as a difficult task for non‐expert users. Furthermore, user queries are often characterized by semantic aspects not directly managed by traditional spatial databases or GIS. Examples of such semantic geospatial queries are the use of implicit spatial relations between objects, or the reference of domain concepts not explicitly represented in data. To handle such queries, we envisage a system that translates natural language queries into spatial SQL statements on a database, thus improving standard GIS with new semantic capabilities. Within this general objective, the contribution of this article is to introduce a methodology to handle semantic geospatial queries issued over a spatial database. This approach captures semantics from an ontology built upon the spatial database and enriched by domain concepts and properties specifically defined to represent the localization of objects. Some examples of the use of the methodology in the urban domain are presented.     

利用自然语言空间关系的空间查询方法研究

首先,从空间认知语言学的角度研究了空间关系自然语言描述中谓词、量词等的分类方法;然后,在分析空间查询语言特点的基础上建立了自然语言空间查询的4类句法模式;最后,设计了一个基于自然语言空间关系的空间查询系统框架,并进行了初步实现。     

MapXtreme for Java 开发中QureyBuilder接口的应用

采用SQL查询语句来定义图层,是应用开发的核心和技术难点之一。QureyBuilder接口技术是数据库类型数据源开发中至关重要的技术环节之一。本文对MapXtreme Java中SQL查询定义图层的方式进行了分析,介绍了QureyBuilder接口应用机理,并给出了完整的QureyBuilder接口应用实例。