Rule effectiveness in rule-based systems: A credit scoring case study

Rule-based systems may sometimes grow very large, making their acceptance by users and their maintenance quite problematic. One therefore needs to make rule-bases as compact as possible. The classical definition of rule redundancy in the literature is based upon logic and graph theory. Another, complementary, view of redundancy is proposed here. The suggested approach is based on the contribution of individual rules to the overall system’s accuracy.

It is shown here, though an analysis of a real-world credit scoring rule-based system, that by taking into account system’s accuracy, one can sometimes significantly reduce the size of a rule-base; even one which is already free from logic-related abnormalities. The approach taken here is not proposed as a substitution to classical logic and graph-based methods. Rather, it complements them.     

On-board preventive maintenance: a design-oriented analytic study for long-life applications

With respect to the long-life missions associated with NASA’s X2000 Advanced Deep-Space System Development Program, reliability implies a system’s continuous operation for many years in an unsurveyed radiation-intense environment. Further, the stringent constraints on the mass of a spacecraft and the power on-board create unprecedented challenges on the means for achieving the ultra-high mission reliability. In this paper, we present an approach to on-board preventive maintenance which rejuvenates a system by letting system components rotate between on-duty and off-duty shifts, slowing down a system’s aging process and thus enhancing mission reliability. By exploiting nondedicated system redundancy, hardware and software rejuvenation are realized simultaneously without significant performance penalty. Our design-oriented analysis confirms a potential for significant gains in mission reliability from on-board preventive maintenance and provides to us useful insights about the collective effect of age-dependent failure behavior, residual mission life, risk of unsuccessful maintenance and maintenance frequency on mission reliability.     

Taxonomy for characterizing ensemble methods in classification tasks: A review and annotated bibliography

Ensemble methodology, which builds a classification model by integrating multiple classifiers, can be used for improving prediction performance. Researchers from various disciplines such as statistics, pattern recognition, and machine learning have seriously explored the use of ensemble methodology. This paper presents an updated survey of ensemble methods in classification tasks, while introducing a new taxonomy for characterizing them. The new taxonomy, presented from the algorithm designer’s point of view, is based on five dimensions: inducer, combiner, diversity, size, and members’ dependency. We also propose several selection criteria, presented from the practitioner’s point of view, for choosing the most suitable ensemble method.     

基于分布式环境一种动态依赖关系发现的方法

本文提出了一种为分布式应用动态构造依赖性模型的方法。这个方法通过对系统进行主动干扰来获得建模的先验知识,然后基于贝叶斯网络构造方法,对分布式应用的组件间关系建立依赖性模型。和传统的被动建模技术不同的是,这种主动方法不需要事先对系统细节充分了解,它通过在运行环境中部署探针,捕捉和测量与部署的主动干扰相关的系统反馈,通过机器学习的方法识别分布式应用中构件间的动态调用的依赖关系,为分布式应用建立动态运行过程中的依赖性模型。动态建立的依赖性模型可用于分布式应用的运行时管理,用于分布式应用执行过程中的故障定位和恢复,对于分布式应用自主计算环境的实现,提供一种实用的方法。     

图依赖研究与应用综述

图依赖是用于解决图数据的数据一致性问题的数据质量规则。基于图依赖提升数据一致性的过程通常分为图依赖定义与形式化、图依赖自动挖掘、基于图依赖的数据一致性提升三步。介绍了针对数据一致性的图依赖理论,并根据拓展类型将图依赖分为基于结构约束拓展、基于语义约束拓展和基于外部约束拓展的图依赖;综述并对比了从图数据中自动挖掘图依赖及其拓展的算法;分析了应用图依赖提高数据一致性的研究现状;总结了当前研究中仍存在的问题,并依据问题展望了图依赖在数据质量领域的应用前景。     

Criteria selection and analysis for single machine dynamic on-line scheduling with multiple objectives and sequence-dependent setups

Fertile opportunities exist for research involving dynamic and stochastic scheduling with multiple conflicting objectives and sequence-dependent setups as little has been reported in the literature to date. This research focuses on understanding and identifying the criteria that could be combined into a single rule using the linear weighted aggregation approach to consider the contradicting needs of cycle time and delivery accuracy. Eight dispatching criteria are compared and evaluated using discrete event simulation. In most studies, the basic concept is to combine different dispatching criterion that performs the best in each objective into a single rule but this may be insufficient. Simulation results show that it is necessary to take into account not only the criterion’s strength in optimizing a performance objective but also the degree of trade-off it has on the other conflicting performance objectives of interest. A correlation analysis of the objectives used is also presented.     

OOPPS: an object-oriented process planning system

This paper presents an object-oriented approach to the development of a generative process planning system—Object-Oriented Process Planning Systems (OOPPS). The system consists of three functional modules: object-oriented product model (OOPM) module, object-oriented manufacturing facility model (OOMFM) module, and object-oriented process planner (OOPP). The OOPM has a hierarchical structure with six classes of objects, class Products, class SubAssembly, class Part, class CSGTree, class Solid and class Feature. It can represent a product with all detailed information. The OOMFM is used to represent a cellular manufacturing system including machine cells, machine tools fixtures, and cutting tools. The OOPP generates process plans for parts using a multi-level hierarchical planning approach with four levels: cell-level planning, machine-level planning, fixture-level planning and tool-level planning. At the cell-level, all required operations are determined based on the feature specifications. Machine cells are then selected, based on the selected operations, for the minimal inter-cell movement. The machine-level planning selects the machines within the chosen cells. The fixtures are also selected on the selected machines. At the fixture-level planning, part setups and their sequences are selected. Finally, the tool-level planning determines all details for the process plans. An automated progamming system was also developed to link the OOPP to Smart CAM to generate CNC programs. An example has been used to illustrate the approach.     

BP''s multi-enterprise asset management system

BP is one of the largest energy companies in the world with 2003 revenues of $233 billion. In this paper, we analyse its use of an innovative ‘multi-enterprise asset management system’ that supports and enables the asset management strategy of BP's exploration and production division on the UK continental shelf (UKCS). The analysis focuses on how BP connects its business processes with over 1500 suppliers to co-ordinate the maintenance, operation and repair of specialised exploration and production equipment. The systems strategy is novel because it takes the enterprise computing concept and implements it across organisational boundaries—hence the term ‘multi-enterprise system’. This use of a shared system with all of its suppliers is distinctive from the most common way of connecting with economic partners which is to use shared data systems based on common data standards and communication technologies such as EDI and more recently XML-based systems within vertical industries such as RosettaNet. The design of the multi-enterprise system is based on a sophisticated business process management system called Maximo and this is used to illustrate the systems design aspect of the overall information system in the broader contexts of business strategy and information technology infrastructure.     

A measurement method for sizing the structure of UML sequence diagrams

ContextThe COSMIC functional size measurement method on UML diagrams has been investigated as a means to estimate the software effort early in the software development life cycle. Like other functional size measurement methods, the COSMIC method takes into account the data movements in the UML sequence diagrams for example, but does not consider the data manipulations in the control structure. This paper explores software sizing at a finer level of granularity by taking into account the structural aspect of a sequence diagram in order to quantify its structural size. These functional and structural sizes can then be used as distinct independent variables to improve effort estimation models.ObjectiveThe objective is to design an improved measurement of the size of the UML sequence diagrams by taking into account the data manipulations represented by the structure of the sequence diagram, which will be referred to as their structural size.MethodWhile the design of COSMIC defines the functional size of a functional process at a high level of granularity (i.e. the data movements), the structural size of a sequence diagram is defined at a finer level of granularity: the size of the flow graph of their control structure described through the alt, opt and loop constructs. This new measurement method was designed by following the process recommended in Software Metrics and Software Metrology (Abran, 2010).ResultsThe size of sequence diagrams can now be measured from two perspectives, both functional and structural, and at different levels of granularity with distinct measurement units.ConclusionIt is now feasible to measure the size of functional requirements at two levels of granularity: at an abstract level, the software functional size can be measured in terms of COSMIC Function Point (CFP) units; and at a detailed level, the software structural size can be measured in terms of Control Structure Manipulation (CSM) units. These measures represent complementary aspects of software size and can be used as distinct independent variables to improve effort estimation models.     

Behavioural equivalence in simulation modelling

Discrete event simulation has grown up as a practical technique for estimating the quantitative behaviour of systems, where direct measurement is undesirable or impractical. It is also used to understand the detailed functional behaviour of such systems. Its theory is largely that of experimental science, centering on statistical approaches to validation, rather than on the verification of detailed behaviour. On the other hand, much work has been done on understanding and proving functional properties of systems, using techniques of formal specification and concurrency modelling. This article presents an approach to understanding equivalence of behaviour of discrete event simulation models, using a technique from the concurrency world, Milner’s Calculus of Communicating Systems (CCS). This yields a significant advance over the main previous work, Schruben and Yücesan’s simulation graphs. CCS allows for the use of observational equivalence, which can capture a more flexible, behavioural notion of equivalence than the structural equivalence defined there.A common framework based on the process view of models is constructed, using a hierarchical graphical modelling language (Extended Activity Diagrams). This language is shown to map onto both the major constructs of the DEMOS discrete event simulation language and the corresponding CCS models. A graphically driven tool based on such a framework is presented, which generates both types of models. Using the CCS model, behavioural equivalences and differences in simulation models are demonstrated.