Midwinters, end games, and body parts: a classification of part-whole relations

This paper deals with the conceptual part-whole relation as it occurs in language processing, visual perception, and general problem solving. One important long-term goal is to develop a naive or common sense theory of the mereological domain, that is the domain of parts and wholes and their relations. In this paper, we work towards such a theory by presenting a classification of part-whole relations that is suitable for different cognitive tasks and give proposals for the representation and processing of these relations. In order to be independent of specific tasks like language understanding or the recognition of objects, we use structural properties to develop our classification.The paper starts with a brief overview of the mereological research in different disciplines and two examples of the role of part-whole relations in linguistics (possessive constructions) and knowledge processing (reasoning about objects). In the second section, we discuss two important approaches to mereological problems: the "Classical Extensional Mereology" as presented by Simons and the meronymic system of part-whole relations proposed by Winston, Chaffin and Hermann. Our own work is described in the third and last section. First, we discuss different kinds of wholes according to their inherent compositional structure; complexes, collections, and masses. Then partitions induced by or independent of the compositional structure of a whole are described, accompanied by proposals for their processing.     

Implementation of meronymic (part-whole) inheritance for semantic networks

Studies in the field of cognitive science have demonstrated the existence of a taxonomy of meronymic (part-whole) relations, each of which corresponds to a meaning of the English phrase ‘part of’. It is proposed here that equivocations between meronymic and other types of semantic relations can be a source of ambiguity during knowledge elicitation, and are, in some part, responsible for the confusing plethora of types of is_a link to be found in current knowledge-based system environments. Based on the taxonomy of part-whole relations, a general-purpose implementation, and applications of meronymic inheritance for a Rule Based Frame System, are presented.     

Identifying object isomerism in multidatabase systems

In a multidatabase system that consists of object databases, the same real-world entity can be stored as objects in different databases with incompatible object identifiers. How to identify and integrate these objects representing the same entities such that (a) object duplication in the query result can be avoided, (b) information for the entity can be gathered, and (c) the specialization of multiple classes can be built is an important issue to provide a well structured global object schema and a more informative query result. In this paper, we extend our results on probabilistic query processing and joining relations on incompatible keys to solve the problem. Various data and schema conflicts such as missing data, inconsistent data and domain mismatch which may exist in classes from different databases are considered in the process of identification.Recommended by: Amit Sheth     

从Web中获取部分整体关系

随着互联网的迅速发展,Web逐步成为知识获取的重要资源。部分整体关系获取是知识获取中的重要组成部分。该文提出了一种利用搜索引擎从Web中获取部分整体关系的方法。首先构造一种基于部分整体关系分类的意图查询,利用意图查询可以有针对性地从Web中获取尽可能多的包含部分整体关系语料。然后根据网页中的HTML标记和意图查询的格式过滤语料,并从中抽取候选部分整体关系,最后基于部分整体关系在自然语言表述中的特点和汉语的构词规律,提出用于验证候选部分整体关系的度量标准。实验结果表明,该方法取得了较高的准确率和F值。在前20个结果中准确率为86%,最优F值为64%。     

On identifying and evaluating object architectures for real-time applications

This paper presents a methodology and tool support for the development of distributed real-time object-oriented systems, focusing on industrial automation applications. At system level, two different kinds of classes/objects are recognized: (i) application domain objects, which map directly to concepts and components of the problem domain, and (ii) design objects that are related to the functional requirements that the system have to meet. The approach provides a method for automatic identification of possible design objects architectures. A set of quality and testability metrics is proposed to evaluate the generated architectures, allowing the identification of the ‘best-fitted’ one.     

Part-whole relations in object-centered systems: An overview

Knowledge bases, data bases and object-oriented systems (referred to in the paper as Object-Centered systems) all rely on attributes as the main construct used to associate properties to objects; among these, a fundamental role is played by the so-called part-whole relation. The representation of such structural information usually requires particular semantics together with specialized inference and update mechanisms, but rarely do current modelling formalisms and methodologies give it a specific, ‘first-class’ dignity. The main thesis of this paper is that the part-whole relation cannot simply be considered as an ordinary attribute: its specific ontological nature requires to be understood and integrated within data-modelling formalisms and methodologies. On the basis of such an ontological perspective, we survey the conceptual modelling issues involving part-whole relations, and the various modelling frameworks provided by knowledge representation and object-oriented formalisms.     

Approximate analysis of binary topological relations between geographic regions with indeterminate boundaries

 The development of formal models of spatial relations is a topic of great importance in spatial reasoning, geographic information systems (GIS) and computer vision, and has gained much attention from researchers across these research areas during the past two decades. In recent years significant achievements have been made on the development of models of spatial relations between spatial objects with precisely defined boundaries. However, these models cannot be directly applied to spatial objects with indeterminate boundaries which are found in many applications in geographic analysis and image understanding. This article develops a method for approximately analyzing binary topological relations between geographic regions with indeterminate boundaries based upon previous work on topological spatial relations and fuzzy sets. In addition, examples are given to demonstrate the method and related concepts. It is shown that the eight binary topological relations between regions in a two-dimensional space can be easily determined by the method.     

空间方向关系描述模型及其GIS应用分析

空间方向关系建模是一个属于空间认知范畴的研究问题。近20年来,该研究问题受到来自计算机、人工智能、机器人以及地理信息科学等领域的众多学者们关注,提出了许多形式化描述模型。首先阐述了方向关系描述框架、表达形式、基本性质以及影响方向关系描述的主要因素。然后,重点回顾及评价了一些较有代表性的方向关系建模方法及其在GIS空间查询、分析、推理中的应用,指出了其中存在的一些主要问题。最后,展望了方向关系模型及其应用中有待进一步研究的若干相关工作。     

基于并列结构的部分整体关系获取方法

部分整体关系是一种基础而重要的语义关系,从文本中自动获取部分整体关系是知识工程的一项基础性研究课题。该文提出了一种基于图的从Web中获取部分整体关系的方法,首先利用部分整体关系模式从Google下载语料,然后用并列结构模式从中匹配出部分概念对,据此形成图,用层次聚类算法对该图进行自动聚类,使正确的部分概念聚集在一起。在层次聚类基础上,我们挖掘并列结构的特性、图的特点和汉语的语言特点,采用惩罚逗号边、去除低频边、奖励环路、加重相同后缀和前缀等5种方法调整图中边的权重,在不损失层次聚类的高准确率条件下,大幅提高了召回率。     

Bottom-up construction of ontologies

Presents a particular way of building ontologies that proceeds in a bottom-up fashion. Concepts are defined in a way that mirrors the way their instances are composed out of smaller objects. The smaller objects themselves may also be modeled as being composed. Bottom-up ontologies are flexible through the use of implicit and, hence, parsimonious part-whole and subconcept-superconcept relations. The bottom-up method complements current practice, where, as a rule, ontologies are built top-down. The design method is illustrated by an example involving ontologies of pure substances at several levels of detail. It is not claimed that bottom-up construction is a generally valid recipe; indeed, such recipes are deemed uninformative or impossible. Rather, the approach is intended to enrich the ontology developer's toolkit     

系统辨识的最优输入设计（OID）综述

本文分时域法和频域法综述了动态系统辨识的最优输入设计。首先回顾了最优输入设计的发展历史,接着主要就七十年代末及八十年代的主要研究成果进行了评述。同时,还介绍了最优输入设计的实际应用研究。最后,本文还对今后的发展方向提出了建议。     

On the improvement of combined fuzzy topological and directional relations information

Concept of combined extraction of topological and directional relations information developed by Zahzah et al. [1] by employing the Allen's temporal relations in 1D spatial domain was improved by Matsakis and Nikitenko [2]. This latter algorithm has high computational complexity due to its limitations of object approximation and segment fuzzification.In this paper, fuzzy Allen relations are used to define the fuzzy topological and directional relations information between different objects. Some extended results of Salamat and Zahzah [3] are discussed. Polygonal object approximation allows us to use fuzzy operators and this approach reduces computational complexity of the method for computing the combined topological and directional relations. To validate the method, some experiments are tested giving satisfactory and promising results. Affine transformation are depicted, these properties will be helpful for using the method in other areas of image analysis such as object tracking.     

从Web获取部分整体关系语料的方法

部分整体关系获取是知识获取中的重要组成部分。Web逐步成为知识获取的重要资源之一。搜索引擎是从Web中获取部分整体关系知识的有效手段之一,我们将Web中包含部分整体关系的检索结果集合称为部分整体关系语料。由于目前主流搜索引擎尚不支持语义搜索,如何构造有效的查询以得到富含部分整体关系的语料,从而进一步获取部分整体关系,就成为一个重要的问题。该文提出了一种新的查询构造方法,目的在于从Web中获取部分整体关系语料。该方法能够构造基于语境词的查询,进而利用现有的搜索引擎从Web中获取部分整体关系语料。该方法在两个方面与人工构造查询方法和基于语料库查询构造查询方法所获取的语料进行对比,其一是语料中含有部分整体关系的语句数量;二是从语料中进一步获取部分整体关系的难易程度。实验结果表明,该方法远远优于后两者。     

A framework for integrating multi-accuracy spatial data in geographical applications

In recent years the integration of spatial data coming from different sources has become a crucial issue for many geographical applications, especially in the process of building and maintaining a Spatial Data Infrastructure (SDI). In such context new methodologies are necessary in order to acquire and update spatial datasets by collecting new measurements from different sources. The traditional approach implemented in GIS systems for updating spatial data does not usually consider the accuracy of these data, but just replaces the old geometries with the new ones. The application of such approach in the case of an SDI, where continuous and incremental updates occur, will lead very soon to an inconsistent spatial dataset with respect to spatial relations and relative distances among objects. This paper addresses such problem and proposes a framework for representing multi-accuracy spatial databases, based on a statistical representation of the objects geometry, together with a method for the incremental and consistent update of the objects, that applies a customized version of the Kalman filter. Moreover, the framework considers also the spatial relations among objects, since they represent a particular kind of observation that could be derived from geometries or be observed independently in the real world. Spatial relations among objects need also to be compared in spatial data integration and we show that they are necessary in order to obtain a correct result in merging objects geometries.     

Making sense of the abstraction hierarchy in the power plant domain

The paper discusses the abstraction hierarchy proposed by Rasmussen [(1986) Information processing and human-machine interaction, North-Holland] for design of human-machine interfaces for supervisory control. The purpose of the abstraction hierarchy is to represent a work domain by multiple levels of means-end and part-whole abstractions. It is argued in the paper that the abstraction hierarchy suffers from both methodological and conceptual problems. A cluster of selected problems are analyzed and illustrated by concrete examples from the power plant domain. It is concluded that the semantics of the means-end levels and their relations are vaguely defined and therefore should be improved by making more precise distinctions. Furthermore, the commitment to a fixed number of levels of means-end abstractions should be abandoned and more attention given to the problem of level identification in the model-building process. It is also pointed out that attempts to clarify the semantics of the abstraction hierarchy will invariably reduce the range of work domains where it can be applied.     

Types, structures and theories in NKI

The National Knowledge Infrastructure (NKI) is a multi-domain knowledge base. The classical type theory is no longer appropriate to describe every kind of object in multi-domains, such as artifacts, natural or micro objects. Three different kinds of type theories are defined: the classical, atomic and pseudo type theories; in the classical type theory, two new type constructors are defined: setm and ∨, to describe the types of sets of all the elements of the types and unions of two sets of different types, respectively. The structures and categories in the type theory are defined, and the sub-structures and homomorphic structures are used to describe the part-of relations that give the algebraic specifications for the natural objects and the part-of relations between the natural objects, micro objects and artifacts.     

The duality of fault-tolerant system structures

An examination of the structure of fault-tolerant systems incorporating backward error recovery indicates a partitioning into two broad classes. Two canonical models, each representing a particular class of systems, have been constructed. The first model incorporates objects and actions as the entities for program construction whereas the second model employs communicating processes and conversations. Applications in areas such as office information and banking systems are typically described and built in terms of the first model whereas applications in the area of process control are usually described and built in terms of the second model. The paper claims that the two models are duals of each other and presents arguments and examples to substantiate this claim. It will be shown that the techniques that have been developed within the context of one model turn out to have interesting and hitherto unexplored duals in the other model.     

Bridging objects and relations: a mediator for an OO front-end to RDBMSs

An object-oriented paradigm is established as the leading approach for developing non-traditional applications, such as GIS or multimedia systems. On the other hand, relational databases have dominated the area of data processing in the past decade. These two trends motivate the research on integrating OO applications with relational databases. This paper presents our approach to the symbiosis of the OO and relational data models, which is built into GinisNT, a scalable OO GIS framework based on an OO-to-relational mapping algorithm. The mapping algorithm transforms classes and objects into relations and tuples, and vice versa, instantiates objects from relational databases. The methodology presented here is extremely efficient, as has been proved by a number of applications developed in GinisNT, and is at the same time cost efficient, as it builds upon existing platforms.