A multi-level approach to modeling language extension in the Enterprise Systems Domain

As the number and diversity of technologies involved in building enterprise systems continues to grow so does the importance of modeling tools that are able to present customized views of enterprise systems to different stakeholders according to their needs and skills. Moreover, since the range of required view types is continuously evolving, it must be possible to extend and enhance the languages and services offered by such tools on an ongoing basis. However, this can be difficult with today׳s modeling tools because the meta-models that define the languages, views and services they support are usually hardwired and thus not amenable to extensions. In practice, therefore, various workarounds have to be used to extend a tool׳s underlying meta-model. Some of these are built into the implemented modeling standards (e.g. UML 2, BPMN 2.0 and ArchiMate 2.0) while others have to be applied by complementary, external tools (e.g. annotation models). These techniques not only increase accidental complexity, they also reduce the ability of the modeling tool to ensure adherence to enterprise rules and constraints. In this paper we discuss the strengths and weaknesses of the various approaches for language extension and propose a modeling framework best able to support the main extension scenarios currently found in practice today.     

Integrated parametric multi-level information and numerical modelling of mechanised tunnelling projects

This paper presents a concept for parametric modelling of mechanized tunnelling within a state of the art design environment, as the basis for design assessments for different levels of details (LoDs). To this end, a parametric representation of each system component (soil with excavation, tunnel lining with grouting, Tunnel Boring Machine (TBM) and buildings) is developed in an information model for three LoDs (high, medium and low) and used for the automated generation of numerical models of the tunnel construction process and soil-structure interaction. The platform enables a flexible, user-friendly generation of the tunnel structure for arbitrary alignments based on predefined structural templates for each component, supporting the design process and at the same time providing an insight into the stability and safety of the design. This model, with selected optimal LoDs for each component, dependent on the objective of the analysis, is used for efficient design and process optimisation in mechanized tunnelling. Efficiency and accuracy are further demonstrated through an error-free exchange of information between Building Information Modelling (BIM) and the numerical simulation and with significantly reduced computational effort. The interoperability of the proposed multi-level framework is enabled through the use of an efficient multi-level representation context of the Industry Foundation Classes (IFC). The results reveal that this approach is a major step towards sensible modelling and numerical analysis of complex tunnelling project information at the early design stages.     

New perspectives in ontological analysis: Guidelines and rules for incorporating modelling languages into UEML

Ontological analysis of modelling languages has been mainly used for evaluating quality of modelling language w.r.t. one specific upper ontology. Generally speaking this evaluation has been done by identifying the coverage of the modelling language constructs w.r.t. the ontology and vice-versa. However, a quite limited support has been developed for performing the ontological analysis task. Specifically, various ontologies used for ontological analysis are not associated to a machine readable format; the coverage of modelling language constructs is mostly provided by informal tables mapping one construct on to one ontological concept; the way in which this coverage task is undertaken is poorly specified (resulting in distinct results for distinct experts involved), and finally, preventing any ontology enrichment for dealing with some specialised language constructs. This limited support also prevents application of ontological analysis outcomes to problems and domains dealing with interoperability, integration and integrated usage of enterprise and IS models, which is today one of the key aspects for making interoperable, maintainable and evolvable inter and intra enterprise software systems. The paper provides an overview of the Unified Enterprise Modelling Language (UEML) approach, which introduces advanced support to ontological analysis of modelling languages. The paper is specifically focused on the task of ontological analysis of modelling languages (named incorporation of modelling languages) by introducing and explaining several guidelines and rules for driving the task: therefore, not all the aspects of the UEML approach will be discussed through the paper. The guidelines and rules are illustrated by incorporation of three selected modelling constructs from IDEF3, a well known language for specifying enterprise processes.     

融合元数据及隐式反馈信息的多层次联合学习推荐方法

基于隐式反馈信息的推荐是目前推荐系统领域的重要方法,能在一定程度上解决显式信息难以获得的问题。但由于隐式数据本身的特点单纯利用隐式反馈信息往往难以获取较好的推荐性能。针对此问题,本文提出一种融合元数据及隐式反馈信息的多层次深度联合学习（Multi-level Deep Joint Learning,简称MDJL）的推荐方法。它利用双深度神经网络共同学习,其中一个网络利用隐式反馈学习用户及项目个体个性化关系,另一个网络利用元数据学习高层次群体共性化关系,从而有效地表达用户偏好,使MDJL框架在个体及群体因素间达到平衡。实验结果表明,MDJL推荐算法在MovieLens 100K和MovieLens 1M两个公开数据集上均表现出更优越的推荐性能。     

Automated text classification of near-misses from safety reports: An improved deep learning approach

Examining past near-miss reports can provide us with information that can be used to learn about how we can mitigate and control hazards that materialise on construction sites. Yet, the process of analysing near-miss reports can be a time-consuming and labour-intensive process. However, automatic text classification using machine learning and ontology-based approaches can be used to mine reports of this nature. Such approaches tend to suffer from the problem of weak generalisation, which can adversely affect the classification performance. To address this limitation and improve classification accuracy, we develop an improved deep learning-based approach to automatically classify near-miss information contained within safety reports using Bidirectional Transformers for Language Understanding (BERT). Our proposed approach is designed to pre-train deep bi-directional representations by jointly extracting context features in all layers. We validate the effectiveness and feasibility of our approach using a database of near-miss reports derived from actual construction projects that were used to train and test our model. The results demonstrate that our approach can accurately classify ‘near misses’, and outperform prevailing state-of-the-art automatic text classification approaches. Understanding the nature of near-misses can provide site managers with the ability to identify work-areas and instances where the likelihood of an accident may occur.     

An ontology modelling perspective on business reporting

In this paper, we discuss the motivation and the fundamentals of an ontology representation of business reporting data and metadata structures as defined in the eXtensible business reporting language (XBRL) standard.     

Incremental constraint modelling in a feature modelling system

The techniques of constraint propagation have recently been successfully applied to feature-based design. Because of their speed, constraint propagation methods allow incremental design and rapid local modifcations of the part. However, cyclic constraints cause serious problems to current constraint propagation algorithms. Variational geometric design systems can, in principle, manage these cases. Unfortunately, this typically requires complete re-evaluation of the underlying set of constraint equations, making the method unsuitable for interactive use. The proposed system aims to localize the problem of constraint solving and maintenance. The constraint graph of the part or assembly is divided into several independent partial graphs, subsystems. Afterwards, each subsystem is handled separately using a selected constraint solving technique for the subsystem.     

基于Deep Belief Nets的维吾尔语句子级情感分析

针对维吾尔语句子情感信息：喜、怒、哀、乐和客观五分类任务,提出一种利用深度信念网络模型(Deep Belief Nets, DBN)的深度学习机制进行基于深层语义特征的句子级情感分析方法。该方法通过对维吾尔语情感句及语言特点的深入研究,提取出利于情感分析任务的8项情感特征。为了提高特征对文本语义的表达,将富含词汇深层语义和上下文信息的word embedding特征与情感特征进行融合,作为深度信念网络的输入。利用多层无监督的波尔兹曼机(RBM)训练并提取隐含的深层语义特征,通过有监督的后向传播算法对网络进行微调,进而完成情感分类任务。该方法在维吾尔语句子级情感分类任务中的准确率为83,35%,召回率为84.42%,F值为：83.88%. 实验结果证明,深度学习模型较浅层的学习模型更合适于本文的情感分类任务,对word embedding特征项的引入,有效的提高了情感分类模型的性能.     

In defence of sublinguistics

The paper is a response to the argument presented by Jussi Karlgren that the term sublanguage involves a potentially damaging misunderstanding in human-computer interaction research. The use of the term in the context of a mathematical concept of language as a set is defended. The definition of language in terms of probability and the use of domain context to condition probabilities is suggested as a refinement to this concept.     

Integrating part and assembly modelling

Current modelling systems adequately support either modelling of parts or modelling of assemblies, whereas ideal modelling systems should adequately support both. To achieve this, a new modelling system has been developed, which uses enhanced multiple-view feature modelling. This advanced modelling approach provides specialised interpretations of a product for different development phases, by means of so-called feature views, and ways to keep these interpretations consistent, i.e. to make sure that they all represent the same product. The paper concentrates on the views that support detail design of parts and assembly design of the whole product, and the way these views are related and kept consistent. It describes the features and the tools that can be used to build and maintain the feature models of the views. An example modelling session is given to illustrate the benefits of such integrated modelling.     

CIMOSA modelling processes

Engineering, integrating and managing complex enterprises requires the understanding, and the ability to partition and simplify their operational complexity. Enterprise modelling supports these requirements by providing means for describing process oriented systems and decomposing those into manageable pieces. However, enterprise modelling requires both a common modelling language and a sufficient modelling methodology. The language provides for common understanding on enterprise models across the industrial community. Modelling methodologies will guide users through the rather complex enterprise modelling tasks. Depending on the skills and the tasks of the modelling person, different methodologies will be implemented in the supporting modelling tool. The paper presents both a methodology for the modelling expert and one for the business user. Whereas the modelling expert will be involved in creating new models, structuring the model contents and developing new modelling components, the business user will use process models for decision support. The latter therefore has a need to modify and adapt enterprise models to represent operational alternatives. A methodology for this type of work has to be based on menus. Menus which are created and maintained by the modelling expert. The business user will mostly work with existing process models. He will evaluate process alternatives and will implement the best solution as the new model of his tasks. This mode of operation will thereby provide for automatic update of the models and will keep the models in sync with the changing reality.     

基于特征选择和深度信念网络的文本情感分类算法

由于人类语言的复杂性，文本情感分类算法大多都存在因为冗余而造成的词汇量过大的问题。深度信念网络（DBN）通过学习输入语料中的有用信息以及它的几个隐藏层来解决这个问题。然而对于大型应用程序来说，DBN是一个耗时且计算代价昂贵的算法。针对这个问题，提出了一种半监督的情感分类算法，即基于特征选择和深度信念网络的文本情感分类算法（FSDBN）。首先使用特征选择方法（文档频率（DF）、信息增益（IG）、卡方统计（CHI）、互信息（MI））过滤掉一些不相关的特征从而使词汇表的复杂性降低；然后将特征选择的结果输入到DBN中，使得DBN的学习阶段更加高效。将所提算法应用到中文以及维吾尔语中，实验结果表明在酒店评论数据集上，FSDBN在准确率方面比DBN提高了1.6%，在训练时间上比DBN缩短一半。     

Deep Web接口的自动识别技术研究

获取Deep Web中信息的主要途径是通过在其提供的查询接口上提交查询来实现的,目前大部分的研究以表单内的

标签获得表单内容结构,判断是不是一个Deep Web查询接口。提出了接口块的概念,设计了一种基于页面信息和视觉信息的接口块定位方法,最后将判定接口块是不是Deep Web接口看作是一个模式识别的分类问题,通过抽取适当的表单结构特征,采用C4.5决策树和SVM相结合的分类算法来进行接口块的判定,得到页面中含有的Deep Web查询接口。采用UIUC的TEL-8数据集进行实验,结果表明,该方法的准确率达到了97.30%,具有良好的可行性和实用性。     

Effective strategies of metamodelling of industrial metallurgical processes

The main objective of the metamodelling is replacing the model of analysed process by its simple (with respect to the computation time) approximation. Metamodel gives a significant reduction of computation time of considered process simulation, as well as its further analysis (sensitivity analysis, optimization, etc.). The paper discusses the idea of metamodelling and compares the effectiveness of three techniques: Response Surface Methodology (RSM), Kriging method and Artificial Neural Network (ANN) applied to the benchmark functions. An example of the use of the considered metamodelling techniques in optimization of the problem of laminar cooling of rolled Dual Phase (DP) steel strips is presented. Metamodelling and optimization of a real industrial metal forming problems seems a novel approach in the field of research on Artificial Intelligence and Optimization practical applications.     

Protocol modelling: A modelling approach that supports reusable behavioural abstractions

We describe a behavioural modelling approach based on the concept of a “Protocol Machine”, a machine whose behaviour is governed by rules that determine whether it accepts or refuses events that are presented to it. We show how these machines can be composed in the manner of mixins to model object behaviour and show how the approach provides a basis for defining reusable fine-grained behavioural abstractions. We suggest that this approach provides better encapsulation of object behaviour than traditional object modelling techniques when modelling transactional business systems. We relate the approach to work going on in model driven approaches, specifically the Model Driven Architecture initiative sponsored by the Object Management Group. Communicated by August-Wilhelm Scheer Ashley McNeile is a practitioner with over 25 years of experience in systems development and IT related management consultancy. His main areas of interest are requirements analysis techniques and model execution and in 2001 he founded Metamaxim Ltd. to pioneer new techniques in these areas. He has published and presented widely on object oriented development methodology and systems architecture. Nicholas Simons has been working with formal methods of system specification since their introduction, and has over 20 years experience in building tools for system design, code generation and reverse engineering. In addition, he lectures on systems analysis and design, Web programming and project planning. He is a co-founder and director of Metamaxim Ltd.     

Deep Web爬虫爬行策略研究

如今Web上越来越多的信息可以通过查询接口来获得,为了获取某Deep Web站点的页面用户不得不键入一系列的关键词集。由于没有直接指向Deep Web页面的静态链接,当前大多搜索引擎不能发现和索引这些页面。然而,近来研究表明Deep Web站点提供的高质量的信息对许多用户来说是非常有价值。这里研究了怎样建立起一个有效的Deep Web爬虫,它可以自动发现和下载Deep Web页面。由于Deep Web惟一“入口点”是查询接口,Deep Web爬虫设计面对的主要挑战是怎样对查询接口自动产生有意义的查询。这里提出一种针对查询接口查询自动产生问题的理论框架。通过在实际Deep Web站点上的实验证明了此方法是非常有效的。     

Agile enterprise architecture modelling: Evaluating the applicability and integration of six modelling standards

ContextAgile enterprise architecture artefacts are initially architected at the high-level and the details of those artefacts iteratively evolve in small project increments. There is a need to model agile enterprise architecture artefacts both at the high and low detailed level for a particular context. ArchiMate is relatively a new high-level architecture modelling standard. There is a growing interest amongst organisations in applying ArchiMate for high-level agile enterprise architecture modelling. However, organisations are unsure how to effectively apply ArchiMate at high-level and integrate it with their existing low detailed level modelling standards in practice for supporting end-to-end agile enterprise architecture modelling.ObjectiveThis paper evaluates the applicability and integration of high-level ArchiMate modelling standard with the existing low-level modelling standards such as BPMN (Business Process Model and Notation), UML (Unified Modelling Language), FAML (FAME [Framework for Agent-Oriented Method Engineering] Language), SoaML (Service Oriented Architecture Modelling Language), and BMM (Business Motivation Model).MethodA qualitative questionnaire-based evaluation criteria has been developed based on the well-known and comprehensive The Open Group Architecture Framework (TOGAF). The evaluation criteria has been applied to evaluate the applicability and integration of the selected six modelling standards from the business, application, infrastructure and extension perspectives.ResultsEach modelling standard is different in scope. A single modelling standard usually does not provide the kind of support required by the agile enterprise architecture modelling. Based on the review results, a hybrid enterprise architecture modelling approach is proposed. This paper demonstrates the application of the proposed hybrid approach with the help of an agile enterprise architecture modelling case study.ConclusionIt is concluded that the ArchiMate does not replace the existing low-level modelling standards, rather it can be used in conjunction with low-level modelling standards. This calls for the adoption of hybrid and integrated approach for agile enterprise architecture modelling.     

Operational building blocks for business process modelling

The modelling of business processes is a rather complex, difficult and therefore time-consuming process. The efficiency of this process can be significantly enhanced by employing reusable building blocks, rather than to describe business processes from scratch every time a model is being created or modified. The paper describes a reference model which identifies a set of generic building blocks needed for enterprise modelling and identifies their relationships. Further specialisation of these generic building blocks into a profession and/or industry branch oriented set enables business people to model their processes with a minimum requirement on IT know-how. The CIMOSA concepts and language have been selected as the basis for the reference model. Since CIMOSA is one of the basic concepts recognised in the relevant standardisation work, the presented approach also obeys very closely the relevant European and international standards.     

多层次深度网络融合人脸识别算法*

深度学习模型可以获得更具有鉴别力的人脸特征,提高人脸识别性能.因此,文中结合深度学习思想,提出多层次深度网络融合特征提取模型.在深度子空间基础上,采用“卷积-池化”网络结构,在降低特征维度的同时保留图像纹理信息,并且获得局部转换鲁棒性.同时,利用人脸标定算法获得人脸特征点,并以此划分人脸区域为5个局部人脸块.基于多层次分类策略,利用全局人脸训练全局网络,完成测试样本预分类.利用局部人脸块训练局部网络,在候选类别中完成最终分类.实验表明,结合局部特征与全局特征的模型可以取得较好的识别率,对光照、表情、姿态,遮挡等影响因素具有较好的鲁棒性,并且加入池化层及两步判别的算法可以有效提高识别率.     

基于复杂网络描述的图像深度卷积分类方法

为了在不增加较多计算量的前提下,提高卷积网络模型用于图像分类的正确率,提出了一种基于复杂网络模型描述的图像深度卷积分类方法。首先,对图像进行复杂网络描述,得到不同阈值下的复杂网络模型度矩阵;然后,在图像度矩阵描述的基础上,通过深度卷积网络得到特征向量;最后,根据得到的特征向量进行K近邻（KNN）分类。在ILSVRC2014数据库上进行了验证实验,实验结果表明,所提出的模型具有较高的正确率和较少的迭代次数。