A laboratory study of user characteristics and decision-making performance in end-user computing

A laboratory study using a simulated business setting was conducted to examine relationships among user characteristics, DSS usage patterns, and decision-making performance. MBA-student subjects self-selected into one of two categories: (1) “end-user programmers” who chose to use the DSS modeling language to develop personalized decision models, or (2) “command level users” who used only “canned” models and query language retrievals. The results show that early development of personalized decision models was positively related to the end-user's previous computer experience. The findings also support the contention that end-user programming improves decision-making performance. End-user programmers consistently outperformed the command level users in terms of their firm's stock price, market share, and return on assets.     

INVEX: Investment Advisory Expert System

Abstract: Capital investment is a very important business decision, because it is largely irreversible and usually long-term. We believe that the use of expert systems as a decision aid for interactive investment decision-making offers several advantages to an unassisted human decision-maker, or even to the human/conventional-DSS combination. For example, the expert system could capture valuable information about so called 'hard data', and about the attitude towards some drastic changes in the environment, that are not easy to include in any quantitative method. However, if one expert can play a very important role in the decision-making process, perhaps different experts can be even more valuable, bringing different approaches and somewhat different sets of information to a decision-making situation. Our multiparadigm blackboard framework, called BEST (Blackboard-based Expert Systems Toolkit), allows its user to combine knowledge coming from different experts, and to use different methods/paradigms to capture that knowledge, according to the type of partial problem at hand where each knowledge source is a single paradigm program. In investment decision-making, judgmental investment ranking and selection from expert economists, embedded in a rule-based knowledge source, might be combined with decisions from operational research methods (embedded in a knowledge source that fully respects multicriteria optimization paradigm) and from risk analysis method (embedded in a conventional, procedural knowledge source). When the decisions being combined come from different types of knowledge sources, redundancy is likely to be reduced and the combined decision is likely to be more objective.     

Systematic modeling and model handling for manpower planning systems

The primary purpose of this paper is to show an efficient way of handling models and model data in a decision support system, in which it is usual to consider several variants of a model. The model data primarily consist of model-defining data, but the same approach may be used for the generated results as well. By efficient handling is meant the handling by the computer as well as by the user. For the user it is particularly important that new models can be conveniently defined as variants of existing models. The approach is introduced within the context of a decision support system for manpower planning based on Markov models. In the mean time the same approach has been used for the implementation of other decision support systems and has been found to be more generally applicable.     

Model theory approach to transaction processing system development

This paper presents a new approach, model theory approach, to small and medium scale transaction processing system (TPS) development. A TPS of this paper is an information system designed to process day-to-day business event data at operational level of an organization. The paper is not concerned with data base construction but with transaction processing.

The model theory approach is not a software engineering approach but a systems theory approach. In the approach a model of the target system, which is called a user model, is constructed in set theory using a formal system structure of a TPS. The user model is, then, compiled into an extended Prolog (extProlog) model. The extProlog is an extension of Prolog to meet requirements for management information system development. On compilation a standardized user interface (UI) called internal UI is attached. The extProlog model with the internal UI is, then, executed under control of another standardized UI called an external UI. Implementation is an integral part of the approach. Because the UIs are designed for the formalized (abstract) structure of a TPS, they can be standardized and are provided as black box components to system development. Because a systems developer is required to only build a user model in set theory based on a model theoretic structure in the approach, it is called a model theory approach. Advantages of this approach are that it provides a theoretical structure to information systems development so that systems development can be made an engineering discipline, and facilitates rapid systems development.     

Future trends in model management systems: parallel and distributed extensions

A model management system (MMS) is a computer system which aids in the organization, utilization, and execution of models for decision makers. Currently, MMS frameworks do not provide an estimate of computer resource consumption rates for solvers, or permit alternate computational models. Without an estimate of computer resource consumption rates, the decision maker is unable to evaluate alternative solution approaches to select the most efficient. Even with such an estimate, MMS may fail the analyst if all strategies require computation time which exceeds the deadline. This is largely due to the fact the MMS only permits sequential computation. MMS prototypes limit execution of solvers to sequential, so alternative computational models such as Single Program Multiple Datastream and Multiple Input Multiple Datastream, cannot be used. MMS frameworks should be extended to provide algorithmic complexity of solvers as an evaluation measure and to incorporate parallel computational models. Networks of workstations connected by an Ethernet are an attractive low cost means of providing concurrent execution, and suit the flexible processing needs of MMS.     

A computer-based system for risk management support

The scale and complexity of present day industrial operations involving hazardous substances are such that managers are faced with increasingly demanding decision problems. They must simultaneously consider technological, economic, environmental and sociopolitical factors. As a response to this problem a computer based decision support system is being developed to support risk management activities, with special emphasis given to hazardous chemicals. The IRIMS (Ispra Risk Management Support) system is an attempt to integrate a number of data bases, containing information relevant to risk management, with several existing simulation models which can be used to address problems of environmental assessment, risk analysis and system optimisation. The system is designed to be user friendly and results are displayed through high resolution colour graphics allowing the non specialised user to obtain a “feel” for the problem under investigation. The paper describes the current prototype system, which is geared to handle problems on a European scale, and plans for further developments which will allow more detailed studies on particular countries or regions.     

Ant colony system: a cooperative learning approach to the travelingsalesman problem

This paper introduces the ant colony system (ACS), a distributed algorithm that is applied to the traveling salesman problem (TSP). In the ACS, a set of cooperating agents called ants cooperate to find good solutions to TSPs. Ants cooperate using an indirect form of communication mediated by a pheromone they deposit on the edges of the TSP graph while building solutions. We study the ACS by running experiments to understand its operation. The results show that the ACS outperforms other nature-inspired algorithms such as simulated annealing and evolutionary computation, and we conclude comparing ACS-3-opt, a version of the ACS augmented with a local search procedure, to some of the best performing algorithms for symmetric and asymmetric TSPs     

A hybrid AI/OR decision support tool for backbone communications network design

Much of existing DSS literature views the role of human expertise as primarily that of selecting appropriate formal models for solving a problem or synthesizing sequences thereof. Once a model (or model sequence) is determined, values of decision variables are determined by the model(s) alone. Hence, automated methods for facilitating model selection and synthesis have received considerable attention. However, a single model is often not an accurate abstraction of reality. Also, results from multiple formal models often have to be combined heuristically to obtain practical solutions. Thus, in this paper we explore the premise that human expertise needs to interact with formal models during the process of searching for solution values. Specifically, we describe a hybrid decision support tool for the design of backbone communication networks, a problem recognized as being of considerable complexity. An internal representation of the design process that employs a blackboard, a truth maintainence system and dependency directed backtracking, allows human expertise and formal models to jointly determine decision variable values in a uniform manner. The design tool has been implemented using a combination of Lisp and Fortran. Computational experiments indicate that incorporating human expertise during the search process results in superior complete solutions and added flexibility in satisfying ad hoc requirements. We conjecture that this hybrid search approach is not limited to the telecommunication network design problem and can be extended to other applications.     

基于领域本体的跨系统个性化服务用户模型

目前大多数个性化服务系统采用关键词或关键词向量空间模型表示用户个人偏好,该方法严重制约了用户模型在不同系统间的共享和重用。针对上述问题,提出一种基于领域本体的用户模型,给出其形式化描述,通过引入个性化兴趣度实现用户个人偏好的量化,介绍该模型在群体用户建模中的应用过程。实验表明,该用户模型能准确反映用户兴趣且更新能力和可扩展性较强。     

针对QAP问题的改进型蚁群优化算法研究

本文结合二次分配问题(quadratic assignment problem,QAP)的特点,通过分析传统蚂蚁算法在解决QAP问题时收敛过快,精度不高的缺点,提出一种以ACS(ant colony system)为基础的改进蚁群算法――信息素迭代累积ACS(ACS with accumu-lated pheromone by iteration,ACS_API)。新方法通过对定义启发式信息和信息素更新规则的改进,扩大了搜索空间,从而避免过早收敛,陷入局部最优解中。该算法已应用于QAP标准测试数据,并通过与另外两种先前提出的改进蚂蚁算法(HAS_QAP,ACO_GLS)的比较分析得出了它在算法精度和执行时间上的优势。     

A Knowledge Discovery System with Support for Model Selection and Visualization

The process of knowledge discovery in databases consists of several steps that are iterative and interactive. In each application, to go through this process the user has to exploit different algorithms and their settings that usually yield multiple models. Model selection, that is, the selection of appropriate models or algorithms to achieve such models, requires meta-knowledge of algorithm/model and model performance metrics. Therefore, model selection is usually a difficult task for the user. We believe that simplifying the process of model selection for the user is crucial to the success of real-life knowledge discovery activities. As opposed to most related work that aims to automate model selection, in our view model selection is a semiautomatic process, requiring an effective collaboration between the user and the discovery system. For such a collaboration, our solution is to give the user the ability to try various alternatives and to compare competing models quantitatively by performance metrics, and qualitatively by effective visualization. This paper presents our research on model selection and visualization in the development of a knowledge discovery system called D2MS. The paper addresses the motivation of model selection in knowledge discovery and related work, gives an overview of D2MS, and describes its solution to model selection and visualization. It then presents the usefulness of D2MS model selection in two case studies of discovering medical knowledge in hospital data—on meningitis and stomach cancer—using three data mining methods of decision trees, conceptual clustering, and rule induction.     

Hybrid recommendation approaches for multi-criteria collaborative filtering

Recommender systems are software tools and techniques for suggesting items in an automated fashion to users tailored their preferences. Collaborative Filtering (CF) techniques, which attempt to predict what information will meet a user’s needs from the neighborhoods of like-minded people, are becoming increasingly popular as ways to overcome the information overload. The multi-criteria based CF presents a possibility to provide accurate recommendations by considering the user preferences in multiple aspects and several methods have been proposed for improving the accuracy of these systems. However, the problem of multi-criteria recommendations with a single and overall rating is still considered an optimization problem. In addition, increasing the accuracy in predicting the appropriate items tailored to the users’ preferences is on of the main challenges in these systems. Hence, in this research new recommendation methods using Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Self-Organizing Map (SOM) clustering are proposed to improve predictive accuracy of criteria CF. In this research, SOM enables us to generate high quality clusters of dataset and ANFIS is used for discovering knowledge (fuzzy rules) from users’ ratings in multi-criteria dataset, generating appropriate membership functions (MFs), overall rating prediction and input selection. Using exhaustive search method for input selection, the effective inputs are determined to build the ANFIS models in all generated clusters. Furthermore, new fuzzy-based algorithms, Weighted Fuzzy MC-CF (WFuMC-CF), Fuzzy Euclidean MC-CF (FuEucMC-CF) and Fuzzy Average MC-CF (FuAvgMC-CF), are presented for prediction task in multi-criteria CF. FuEucMC-CF and FuAvgMC-CF algorithms uses the fuzzy-based Euclidian distance and fuzzy-based average similarity, respectively, the WFuMC-CF algorithm uses fuzzy-based user- and item-based prediction in a weighted approach. Experimental results on real-world dataset demonstrate that the proposed hybrid methods remarkably improve the accuracy of multi-criteria CF in relation to the previous methods based on multi-criteria ratings.     

Synchronisation in Trust Management Using Push Authorisation

Traditional trust management authorisation decisions for distributed technologies, are, in general, based on the history of the authorisations/computation to date. We consider this a pull authorisation strategy: the authorisation decision reflects the current and/or past authorisations. In this paper, we examine this pull strategy and propose an alternative form of authorisation in a distributed environment. Instead of 'pulling' the information required for the current authorisation decisions from the past, authorisation decisions are made to specify what will happen in the future. This strategy is called push authorisation. When a push decision is made, its result is pushed to just the relevant protection mechanisms. This approach allows the creation of distributed separation of duties policies, without requiring additional synchronisation between components in the execution. It allows present actions to inform future authorisation decisions, before those decisions must be made.     

异构无线网中面向QoS的马尔可夫选择策略

针对移动终端在异构网络环境下,需要在垂直切换过程中进行网络选择的问题,提出一种面向QoS的马尔可夫选择决策算法,通过对算法模型合理化构建过程与异构环境特点的紧密结合、报酬函数的正确定义与求解,可以为用户选择合适的接入网络,最大程度地满足在异构网络环境中用户QoS的长期效益。仿真结果表明,该算法可以有效提高判决水平,改善业务的QoS。     

Incremental user-interface development for interactive multiobjective optimization

Interactive multiobjective optimization (IMO) is a subfield of multiple criteria decision making. In multiobjective optimization, the optimization problem is formulated with a mathematical model containing several conflicting objectives and constraints depending on decision variables. By using IMO methods, a decision maker progressively provides preference information in order to find the most satisfactory compromise between the conflicting objectives. In this paper, we consider implementation challenges of IMO methods. In particular, we consider what kind of interaction techniques can support the decision making process and information exchange between IMO methods and the decision maker. The implementation of an IMO method called Pareto Navigator is used as an example to demonstrate concrete challenges of interaction design. This paper focuses on describing the incremental development of the user interface for Pareto Navigator including empirical validation by user testing evaluation.     

A blackboard control architecture for model selection and sequencing

Model management systems are computerised systems that facilitate the management of large numbers of decision models used in organizations. Model selection and sequencing in a model management system is the problem of processing a given model base in order to arrive at a sequence of models that can be executed to produce a set of required outputs (goal). Prior solution approaches do not attempt to solve this problem such that the goal is achieved while best meeting the objectives of the user. Instead, research to date has typically provided the first sequence of models which satisfy the goal, without attempting to optimise the objectives of the user. This restricts the applicability of many existing approaches to problems with unique solutions or to situations where users exhibit no preference among the candidate model sequences (i.e. solutions). In many real-world problems, however, multiple solutions may exist and users may prefer a certain solution over the others, based on a variety of criteria such as solution cost, accuracy and so on. In this paper, we present an architecture based on the concept of blackboard control that solves the model selection and sequencing problem while attempting to optimise the objectives of the user. We also discuss the applicability of the proposed approach for solving other problems encountered in the area of model management.     

Public-key encryption with bidirectional keyword search and its application to encrypted emails

A traditional Public-key Encryption scheme with Keyword Search (PEKS) allows multiple senders to encrypt keywords under the public key of a receiver such that the receiver can search on these encrypted keywords using his/her searching secret key. In encrypted email systems, an email user not only needs to search on encrypted emails received from other users, but also needs to search on encrypted emails sent to other users. Motivated by this, the paper proposes a cryptographic method to allow these two types of user (i.e., senders and receivers) to search on encrypted keywords, which is called Public-key Encryption with Bidirectional Keyword Search (PEBKS). We give formal definitions of a PEBKS scheme and its indistinguishable security model to capture the scenario that no adversary can efficiently distinguish two ciphertexts of keywords from each other, even if the adversary can adaptively obtain search trapdoors of many keywords. Specifically, we propose a concrete PEBKS scheme, whose security relies on a standard hard problem, i.e., bilinear Diffie–Hellman problem, in the random oracle model. Finally, we simulate the proposed PEBKS scheme to assess its practicability and convinces that its feasibility to be applied to encrypted email systems.     

User Interface Management Systems for Engineering Applications

Essential features of a User Interface Management System (UIMS) from the perspective of engineering applications' user interface development are described. Three commercially available UIMS are validated against this set of essential features and the results of such a categorised evaluation are reported. The proliferation of user interface design tools necessitated an initialfiltering to select these tools on the basis of their availability on a wide range of platforms currently used by engineers, supplier support for the product and the prototyping tools the UIMS offer. Graphic Modelling System (GMS), TeleUSE and USEIT were selected for their potential utility to engineering applications. An earlier version of this paper was presented at the 1991 European X User Group Annual Conference held at Queens' College, Cambridge, UK. This comparative evaluation is compiled on behalf of the UK Science and Engineering Research Council, for the benefit of the UK academic engineering research community.     

Incremental Beam search

Beam search is a heuristic search algorithm that explores a state-space graph by expanding w most promising nodes at each level (depth) of the graph, where w is called the beam-width which is taken as input from the user. The quality of the solution produced by beam search does not always monotonically improve with the increase in beam-width making it difficult to choose an appropriate beam-width for effective use. We present an algorithm called Incremental Beam Search (IncB) which guarantees monotonicity, and is also anytime in nature. Experimental results on the sliding-tile puzzle, the traveling salesman, and the single-machine scheduling problems show that IncB significantly outperforms basic monotonic methods such as iterative widening beam search as well as some of the state-of-the-art anytime heuristic search algorithms in terms of the quality of the solution produced at the end as well as the anytime performance.     

Thomas: building Bayesian statistical expert systems to aid in clinical decision making.

Knowledge-based system for classical statistical analysis must separate the task of analyzing data from that of using the results of the analysis. In contrast, a Bayesian framework for building biostatistical expert system allows for the integration of the data-analytic and decision-making tasks. The architecture of such a framework entails enabling the system (1) to make its recommendations on decision-analytic grounds; (2) to construct statistical models dynamically; (3) to update a statistical model based on the user's prior beliefs and on data from, the methodological concerns evinced by, the study. This architecture permits the knowledge engineer to represent a variety of types of statistical and domain knowledge. Construction of such systems requires that the knowledge engineer reinterpret traditional statistical concerns, such as by replacing the notion of statistical significance with that of a pragmatic clinical threshold. The clinical user of such a system can interact with the system at a semantic level appropriate to her fund of methodological knowledge, rather than at the level of statistical details. We demonstrate these issues with a prototype system called THOMAS which helps a physician decision maker interpret the results of a published randomized clinical trial.