Evaluation of inductive logic programming for information extraction from natural language texts to support spatial data recommendation services

In this article we analyze a well-known and extensively researched problem: how to find all datasets, on the one hand, and on the other hand only those that are of value to the user when dealing with a specific spatially oriented task. In analogy with existing approaches to a similar problem from other fields of human endeavor, we call this software solution ‘a spatial data recommendation service.’ In its final version, this service should be capable of matching requests created in the user's mind with the content of the existing datasets, while taking into account the user's preferences obtained from the user's previous use of the service. As a result, the service should recommend a list of datasets best suited to the user's needs. In this regard, we consider metadata, particularly natural language definitions of spatial entities, a crucial piece of the solution. To be able to use this information in the process of matching the user's request with the dataset content, this information must be semantically preprocessed. To automate this task we have applied a machine learning approach. With inductive logic programming (ILP) our system learns rules that identify and extract values for the five most frequent relations/properties found in Slovene natural language definitions of spatial entities. The initially established quality criterion for identifying and extracting information was met in three out of five examples. Therefore we conclude that ILP offers a promising approach to developing an information extraction component of a spatial data recommendation service.     

中文文本的地理命名实体标注

通过文本中地理信息的语义解析,可以帮助人们深入理解空间认知和空间语言的表达规律,解决自然语言与地理信息系统（GIS）之间的语义障碍问题,提升GIS空间查询、空间推理、地理信息检索和地理信息服务的智能化水平。制定标注体系和建立标注语料库,能够发现自然语言中地理信息描述的语言结构,建立它们的元数据。本文在分析中文文本和GIS中地理实体描述和表达机制差异的基础上,结合地理命名实体描述的语言特点,制定了中文文本的地理命名实体标注体系和标注规范,并以GATE（General Architecture for Text Engineering）作为标注平台,构建了基于《中国大百科全书中国地理》（简称“GeoCorpus”）的大规模标注语料库,较为有效地解决了当前相关标准和规模化标准数据匮乏的问题。     

Structural setting and tectonic evolution of the Apennine Units of northern Calabria

A new structural–stratigraphic synthesis of the Apennine units of northern Calabria is presented. The Meso-Cenozoic successions are grouped into two tectonic units, named Pollino–Ciagola Unit (PCU) and Lungro–Verbicaro Unit (LVU), comprising terrains formerly attributed to five different tectonic units. FeMg carpholite and blue amphibole record HP–LT metamorphism in the LVU, followed by progressive decompression leading to final greenschist facies re-equilibration during dominantly extensional deformation. Final tectonic emplacement of the LVU over the PCU post-dated the metamorphism of the LVU and was accompanied by intense ductile deformation along zones of strain localisation in footwall rocks. All of the units were later affected by folding and minor thrusting during subsequent Apennine tectonics. To cite this article: A. Iannace et al., C. R. Geoscience 337 (2005).     

基于深度信念网络的地质实体识别方法

地质实体作为地质信息表达的核心要素,对其准确识别是地质文本数据挖掘和应用的重要基础。本文通过分析各种类型文本数据中地质实体信息的描述特点,构建了地质实体信息的标注规范和语料库,设计了基于深度信念网络(Deep Belief Networks)的地质实体识别模型,解决了文本数据中地质实体信息的结构化、规范化处理问题。以矿产资源地质调查报告为实验数据,对本文的地质实体识别方法性能进行了评估分析。结果表明,深度学习模型能够在较小规模语料库的基础上,达到较好的地质实体识别性能。     

面向对象的地理实体时空位置多粒度表达

同一地理实体在不同的时空粒度下会表现出相异的位置动态变化规律。近年来,如何对地理实体在不同时空粒度下的时空位置进行组织和表达成为GIS研究的热点之一。本文基于面向对象的思想,设计了“三级空间”和“0-1位置变化序列”,并由此提出一种地理实体时空位置的多粒度表达方法。在实体时空位置的多粒度描述方面,对于任一地理实体,空间维度上构建一种具有不同空间粒度的“全局—相对—对象”三级空间;时间维度上将不同时段或时刻转换为一系列不同时间粒度的离散时间点。在实体时空位置的多粒度存储组织方面,将地理实体时空位置的变化过程划分为不同阶段,对该实体在不同时间点下的空间位置信息设置不同的存储方式,可合理减少信息冗余。在实体时空位置的多粒度转换方面,提出基于三级空间的递进认知、时间点与时段之间快速转换等策略,初步实现了地理实体时空位置在不同时空粒度下的转换。该方法可有效地描述地理实体在可变时空粒度下的时空位置,为时态GIS和多粒度时空数据库的建立提供参考。     

Characterization of humic substances by advanced solid state NMR spectroscopy: Demonstration of a systematic approach

Characterization of humic substances is challenging due to their structural complexity and heterogeneity. Solid state nuclear magnetic resonance (NMR) is regarded as one of the best tools for elucidating structures of humic substances. The primary solid state NMR technique that has been used so far is the routine ¹³C cross polarization-magic angle spinning (CP-MAS) technique. Although this technique has markedly advanced our understanding of humic substances, the full potential of NMR for characterizing humic substances has yet to be realized. Recent technical developments and applications of advanced solid state NMR have revealed the promise to provide deeper insights into structures of humic substances. In this paper, we summarized and demonstrated the systematic solid state NMR protocol for characterization of humic substances using a humic acid as an example. This protocol included (1) identification of specific functional groups using spectral editing techniques, occasionally assisted by ¹H¹³C two-dimensional heteronuclear correlation (2D HETCOR) NMR, (2) quantification of specific functional groups based on direct polarization-magic angle spinning (DP-MAS) and DP-MAS with recoupled dipolar dephasing, combined with spectral editing techniques, (3) determination of connectivities and proximities of specific functional groups by ¹H¹³C 2D HETCOR or 2D HETCOR combined with spectral editing techniques, and (4) examination of domains and heterogeneities by ¹H¹³C 2D HETCOR with ¹H spin diffusion. We used a soil humic acid as an example to demonstrate how this protocol was applied to the characterization of humic substances step by step. Afterwards, based on typical ¹³C NMR spectra of humic substances we described how we could combine different NMR techniques to identify specific functional groups band by band from downfield to upfield. Finally, we briefly mentioned the potential new NMR techniques that could be developed to enrich the current systematic protocol. This systematic protocol is not only applicable to humic substances but also to other natural organic matter samples.     

GIS中基于角色的图元授权粒度控制研究

对深入到图元的细粒度控制进行了研究,分析了图元授权粒度需要考虑的各个方面,在RBAC理论基础上,对该控制做了全面而细致的设计。实践证明,GIS应用系统中基于角色的图元授权粒度控制对具体工作单位的运作具有真正意义上的实用性。     

面向地理实体的三维模型数据库快速构建技术研究

大规模城市三维建模一直是三维城市建设发展的关键,本文按照基础数据的分类方法,研究地理实体要素中各类模型构建规律,抽象出点、线、面状要素,利用符号匹配、规则构建及三角网构建方式进行批量建模。该方法实现了对地理实体要素三维表达的快速构建,具有较强的实用性。     

三维拓扑关系的基本问题和研究进展

三维拓扑关系是三维空间关系中最重要的空间关系,是三维空间数据表达、查询、分析与应用的理论基础,是目前拓扑关系研究的热点。本文从三维空间实体的定义和描述、三维拓扑关系描述、表达与应用等方面回顾了国内外学术界所取得的研究进展,同时分析了存在的一些基本问题。