DAISY: a decision support design methodology for complex,experience-centered domains

Users at different levels of domain experience have very different needs. For example, a system designed to assist domain novices may frustrate experts and vice-versa. This is one of several challenges specific to building decision support systems for experience-centered domains. A second challenge in working with complex experience-centered domains is that it is hard for non-experts to understand the domain in order to model it. In this paper we present DAISY, the design aid for intelligent support systems. It is a software design methodology for constructing decision support systems in complex, experience-based domains. DAISY address the specialized challenges of these domains by augmenting existing cognitive engineering methodologies. In particular, DAISY provides a method for identifying the specialized needs of users within a specific range of domain experience. Thus, it can help software designers to understand "What does the domain expert need?" or "What does a trained novice need?" To help system designers manage the complexity of modeling unfamiliar experience-centered domains, it provides a tool called a time/activity matrix. To illustrate each of DAISY's steps, we used the development of a decision support system called Fox. Fox assists expert military planners by rapidly generating alternative plans. This is a cognitively difficult, time critical task with life and death consequences     

A framework for explaining reliance on decision aids

This study presents a framework for understanding task and psychological factors affecting reliance on advice from decision aids. The framework describes how informational asymmetries in combination with rational, motivational and heuristic factors explain human reliance behavior. To test hypotheses derived from the framework, 79 participants performed an uncertain pattern learning and prediction task. They received advice from a decision aid either before or after they expressed their own prediction, and received feedback about performance. When their prediction conflicted with that of the decision aid, participants had to choose to rely on their own prediction or on that of the decision aid. We measured reliance behavior, perceived and actual reliability of self and decision aid, responsibility felt for task outcomes, understandability of one's own reasoning and of the decision aid, and attribution of errors. We found evidence that (1) reliance decisions are based on relative trust, but only when advice is presented after people have formed their own prediction; (2) when people rely as much on themselves as on the decision aid, they still perceive the decision aid to be more reliable than themselves; (3) the less people perceive the decision aid's reasoning to be cognitively available and understandable, the less people rely on the decision aid; (4) the more people feel responsible for the task outcome, the more they rely on the decision aid; (5) when feedback about performance is provided, people underestimate both one's own reliability and that of the decision aid; (6) underestimation of the reliability of the decision aid is more prevalent and more persistent than underestimation of one's own reliability; and (7) unreliability of the decision aid is less attributed to temporary and uncontrollable (but not external) causes than one's own unreliability. These seven findings are potentially applicable for the improved design of decision aids and training procedures.     

An ID-based authenticated dynamic group key agreement with optimal round

Group key agreement protocols are crucial for achieving secure group communications.They are designed to provide a set of users with a shared secret key to achieve cryptographic goal over a public network.When group membership changes,the session key should be refreshed efficiently and securely.Most previous group key agreement protocols need at least two rounds to establish or refresh session keys.In this paper,a dynamic authenticated group key agreement(DAGKA) protocol based on identity-based cryptography is presented.By making use of the members’ values stored in previous sessions,our Join and Leave algorithms reduce the computation and communication costs of members.In the proposed protocol,Setup and Join algorithms need one round.The session key can be refreshed without message exchange among remaining users in Leave algorithm,which makes the protocol more practical.Its security is proved under decisional bilinear Diffie-Hellman(DBDH) assumption in random oracle model.     

Design and evaluation of visualization support to facilitate decision trees classification

The loosely coupled relationships between visualization and analytical data mining (DM) techniques represent the majority of the current state of art in visual data mining; DM modeling is typically an automatic process with very limited forms of guidance from users. A conceptual model of the visualization support to DM modeling process and a novel interactive visual decision tree (IVDT) classification process have been proposed in this paper, with the aim of exploring humans’ pattern recognition ability and domain knowledge to facilitate the knowledge discovery process. An IVDT for categorical input attributes has been developed and experimented on 20 subjects to test three hypotheses regarding its potential advantages. The experimental results suggested that, compared to the automatic modeling process as typically applied in current decision tree modeling tools, IVDT process can improve the effectiveness of modeling in terms of producing trees with relatively high classification accuracies and small sizes, enhance users’ understanding of the algorithm, and give them greater satisfaction with the task.     

Structuring group decision making in a web-based environment by using the nominal group technique

The Nominal Group Technique (NGT) is a structured decision-making technique widely used both in industry and academia as a tool to aid in planning and decision-making processes. The main goal of this study was to evaluate the performance of NGT in a web-based environment compared to its traditional counterpart. Comparisons were made along several performance and process-related dimensions. First, the decision process and experimental methodology are introduced. Second, the interface design used during the online sessions is described. Next, the response variables were explored following two difference approaches. The first approach compared the response variables from the traditional and the online setting. The second approach assessed the differences in perceived values before and after the session was conducted. Results revealed that the traditional NGT outperformed the online version in the variables related to the process. However, traditional groups did not significantly differ from online groups with respect to the variables related to the outcomes. Finally, we suggest conditions for enhanced productivity in idea-generating and problem-solving groups by providing some sustaining evidence.     

Ironic efficiency in automation-aided signal detection

Abstract

Decision makers often make poor use of the information provided by an automated signal detection aid; recent studies have found that participants assisted by an automated aid fell well short of best-possible sensitivity levels. The present study tested the generalisability of this finding over varying levels of aid reliability. Participants performed a binary signal detection task either unaided or with assistance from a decision aid that was 60%, 85%, or 96%-reliable. Assistance from a highly reliable aid (85% or 96%) improved discrimination performance, while assistance from a low-reliability aid (60%) did not. Because their ideal strategy is to place less weight on less reliable cues, however, the decision makers’ tendency to disuse the aid became more appropriate as the aid’s reliability declined. Automation-aided efficiency was thus near to optimal when the aid was close to chance but became highly inefficient, ironically, as the aid’s reliability increased.

Practitioner Summary: Investigating operators’ automation-aided information integration strategies allows human factors practitioners to predict the level of performance the operator will attain. Ironically, in an aided signal detection task, performance when assisted by a highly reliable aid is far less efficient than that obtained when assisted by a far less reliable aid.

Abbreviations: OW: optimal weighting; UW: uniform weighting; CC: contingent criterion; BD: best decides; CF: coin flip; PM: probability matching; HDI: highest density interval; MCMC: markov chain monte carlo; HR: hit rate; FAR: false alarm rate     

Heuristic-based information acquisition and decision making among pilots

OBJECTIVE: This research was designed to examine the impact of heuristic-based approaches to the acquisition of task-related information on the selection of an optimal alternative during simulated in-flight decision making. BACKGROUND: The work integrated features of naturalistic and normative decision making and strategies of information acquisition within a computer-based, decision support framework. METHOD: The study comprised two phases, the first of which involved familiarizing pilots with three different heuristic-based strategies of information acquisition: frequency, elimination by aspects, and majority of confirming decisions. The second stage enabled participants to choose one of the three strategies of information acquisition to resolve a fourth (choice) scenario. RESULTS: The results indicated that task-oriented experience, rather than the information acquisition strategies, predicted the selection of the optimal alternative. It was also evident that of the three strategies available, the elimination by aspects information acquisition strategy was preferred by most participants. CONCLUSION: It was concluded that task-oriented experience, rather than the process of information acquisition, predicted task accuracy during the decision-making task. It was also concluded that pilots have a preference for one particular approach to information acquisition. APPLICATION: Applications of outcomes of this research include the development of decision support systems that adapt to the information-processing capabilities and preferences of users.     

Using a cognitive model to generate web navigation support

A computational cognitive model of web navigation is proposed. Based on theories and models of text comprehension and web navigation, the plausibility of the proposed model is discussed. The model was used to generate navigation support and this support was offered to users in real time during their navigation sessions, in two experiments. In the first experiment navigation support was offered in the auditory modality and it had a positive effect on user's task performance, especially for users with low spatial abilities. In the second experiment navigation support was offered in the visual modality and users positively evaluated it. Users navigated in a more structured way, judged the system as more usable, and perceived themselves as less disoriented. Support did also here lead to better task performance. Finally, some aspects concerning further enhancement of the validity of the proposed model and its practical relevance are discussed.     

Visual decision support for business ecosystem analysis

This study comparatively evaluates the effectiveness of three visualization methods (list, matrix, network) and the influence of data complexity, task type, and user characteristics on decision performance in the context of business ecosystem analysis. We pursue this objective using an exploratory study with 14 prototypical users (e.g. executives, analysts, investors, and policy makers). The results show that in low complexity contexts, decision performance between visual representations differ but not substantially. In high complexity contexts, however, decision performance suffers significantly if visual representations are not appropriately matched to task types. Our study makes several theoretical and practical contributions. Theoretically, we extend cognitive fit theory by investigating the impact of business ecosystem task type and complexity. Managerially, our study contributes to the relatively underexplored, but emerging area of the design of business ecosystem intelligence tools and presentation of business ecosystem data for the purpose of decision making. We conclude with future research opportunities.     

Human-automated path planning optimization and decision support

Path planning is a problem encountered in multiple domains, including unmanned vehicle control, air traffic control, and future exploration missions to the Moon and Mars. Due to the voluminous and complex nature of the data, path planning in such demanding environments requires the use of automated planners. In order to better understand how to support human operators in the task of path planning with computer aids, an experiment was conducted with a prototype path planner under various conditions to assess the effect on operator performance. Participants were asked to create and optimize paths based on increasingly complex path cost functions, using different map visualizations including a novel visualization based on a numerical potential field algorithm. They also planned paths under degraded automation conditions. Participants exhibited two types of analysis strategies, which were global path regeneration and local sensitivity analysis. No main effect due to visualization was detected, but results indicated that the type of optimizing cost function affected performance, as measured by metabolic costs, sun position, path distance, and task time. Unexpectedly, participants were able to better optimize more complex cost functions as compared to a simple time-based cost function.     

The effects of a group decision support system on culturally diverse and culturally homogeneous group decision making

Considerable research has been conducted over the last decade on the effects of using Group Decision Support Systems (GDSS) in group decision making. Studies have examined the influence of GDSS on such variables as group task, size, quality of decision, and satisfaction with the process. However, only a few studies have focused on the use of GDSS and its interaction with group composition. This study compared the effects of group decision making, both with and without GDSS technology, on culturally diverse and homogeneous groups. Group performance was measured by the number of non-redundant, realistic ideas and the quality of solution produced in response to posed problems. The GDSS used in the study was Vision Quest. Groups using it, regardless of cultural make up, developed a significantly higher number of non-redundant, realistic ideas than groups that did not. Furthermore, among groups using the GDSS, culturally diverse groups produced a significantly higher number of non-redundant, realistic ideas than homogeneous groups that used the GDSS. Implications for diversity management and group support research are suggested.     

基于改进的代价敏感决策树的网络贷款分类

在网络贷款用户数据集中，贷款成功和贷款失败的用户数量存在着严重的不平衡，传统的机器学习算法在解决该类问题时注重整体分类正确率，导致贷款成功用户的预测精度较低。针对此问题，在代价敏感决策树敏感函数的计算中加入类分布，以减弱正负样本数量对误分类代价的影响，构建改进的代价敏感决策树；以该决策树作为基分类器并以分类准确度作为衡量标准选择表现较好的基分类器，将它们与最后阶段生成的分类器集成得到最终的分类器。实验结果表明，与已有的常用于解决此类问题的算法（如MetaCost算法、代价敏感决策树、AdaCost算法等）相比，改进的代价敏感决策树对网络贷款用户分类可以降低总体的误分类错误率，具有更强的泛化能力。     

基于决策分类熵的决策树构造算法及应用

为了更好地完成金融数据集上的分类挖掘任务,以粗糙集理论为基础提出决策分类熵的概念,进而以属性的决策分类熵为属性分裂度量提出基于决策分类熵的决策树构造算法,并针对过拟合问题提出一种抑制参数来实现树规模的良好控制。实例分析及金融数据集上的实验表明：相比经典的C4.5决策树算法,新算法能够较好地克服其缺点和不足,构建更优的决策树,能够更好地完成分类任务。     

Blind equalization using parallel Bayesian decision feedback equalizer

The purpose of this paper is to propose a new method for blind equalization using parallel Bayesian decision feedback equalizer (DFE). Blind equalization based on decision-directed algorithm, including the previous proposed Chen’s blind Bayesian DFE, cannot give the correct convergence without the suitable initialization corresponding to the small inter-symbol interference. How to find the suitable initialization becomes the key for the correct convergence. Here, the “start” vector with several states is used to obtain several channel estimates which are the initial channel estimates in proposed method. In these initial channel estimates, the best one which has converged toward the correct result in some degree must exist. The decision-directed algorithm for parallel blind Bayesian DFE is purchased from these initial channel estimates respectively. Evaluating the Bayesian likelihood which is defined as the accumulation of the natural logarithm of the Bayesian decision variable, the correct channel estimates corresponding to the maximum Bayesian likelihood can be found. Compared with Chen’s blind Bayesian DFE, the proposed method presents better convergence performance with less computational complexity. Furthermore, the proposed algorithm works satisfactorily even for channel with severe ISI and in-band spectral null, while Chen’s blind Bayesian DFE fails.     

Task constraints and user-system interaction process under personnel decision support

Abstract

An experiment involving a simulated decision support system was carried out to examine the patterns of user-system interaction and decision information utilization under personnel decision support. A computer simulation program of personnel decision support was developed, using actual personnel management data from eight Chinese enterprises, and the process tracing techniques were adopted. Thirty-six subjects (users) participated in this experiment. A 2 × 2 design of task constraints was formulated including two forms of decision information representations (chunking vs. random) and two types of time pressure (3 minutes vs. 1 minute). The results showed that, in interacting with decision support systems, users’ weights of decision information attributes were closely correlated with the types of information search patterns. Under high time-pressure and chunking representation condition, more selective search strategies were adopted with a similar pattern of the sequential search as it was under low time-pressure. The user-system interaction revealed a linear additive process of information utilization. Implications of the results are discussed in relation to the design of effective decision support systems for complex decision situations.     

Feature extraction based on decision boundaries

A novel approach to feature extraction for classification based directly on the decision boundaries is proposed. It is shown how discriminantly redundant features and discriminantly informative features are related to decision boundaries. A procedure to extract discriminantly informative features based on a decision boundary is proposed. The proposed feature extraction algorithm has several desirable properties: (1) it predicts the minimum number of features necessary to achieve the same classification accuracy as in the original space for a given pattern recognition problem; and (2) it finds the necessary feature vectors. The proposed algorithm does not deteriorate under the circumstances of equal class means or equal class covariances as some previous algorithms do. Experiments show that the performance of the proposed algorithm compares favorably with those of previous algorithms     

A decision support system for cooperative transportation planning: Design,implementation, and performance assessment

In this paper, we describe a decision support system for cooperative transportation planning in the German food industry where several manufacturing companies share their fleets to reduce transportation costs. Besides using vehicles of their fleets, there are different outsourcing options offered by logistics service providers, but these are much more expensive. The decision-making kernel of the decision support system is implemented as a multi-agent-system (MAS). The kernel provides a distributed hierarchical algorithm for cooperative transportation planning and an on-line data layer that contains all the information for decision making. We sketch the distributed hierarchical transportation planning algorithm and identity the required software agents. The MAS interacts via web services with a commercial tour planning system that persistently stores the resulting tour plans, orders, and customer data. Moreover, the tour planning system is used to offer graphical user interfaces to interact with the users. The data layer is updated by order and customer data from the ERP systems of the different manufacturing companies. We describe the architecture and the implementation of the MAS and the overall coupling framework. Furthermore, we discuss the simulation-based performance assessment of the resulting decision support system when the system is applied in a rolling horizon setting and present some computational results. The results demonstrate that the MAS approach is appropriate for the cooperative transportation planning domain.     

装备测试性设计辅助决策系统关键技术研究

针对测试性设计的辅助决策工具是装备测试性设计水平提高的关键技术。围绕测试性设计辅助决策系统实现,从装备测试性设计建模、装备测试性设计与分析优化算法、辅助决策系统实现等三个方面展开研究,在分析现有模型的基础上提出了用于测试性设计的信息模型建模思想,并对测试性设计过程中的诊断优化算法进行了研究,最后完成对测试性模型、诊断策略的集成,对系统平台实现的关键技术进行了研究。本研究是实现装备测试性设计辅助决策系统的关键。     

The effect of individual and task characteristics on decision aid reliance

This research explores reliance behaviours of decision-makers using a decision aid. Objective and subjective task characteristics in the form of task complexity and task difficulty, respectively, are examined, along with the effect of the individual characteristic of expertise. A total of 130 subjects (65 novices and 65 experienced practitioners) completed a lab experiment using a decision aid (Insolve-DG) to help them make decisions for two insolvency tasks with differing levels of complexity. The research finds that the objective task characteristic (task complexity) and individual characteristic (expertise) both affect reliance behaviours; however, their effects are fully mediated by the subjective task characteristic (task difficulty). Expertise and task complexity are both associated with the degree of task difficulty experienced by an individual user: increasing task complexity increases task difficulty, and increasing expertise reduces task difficulty. Task difficulty and task complexity are established as different constructs; and importantly it is task difficulty, not task complexity, that ultimately affects reliance.     

Decision aid and predicates in production control

Production Control Systems (PCS) belong to the multilevel hierarchical complex systems. This article presents a four level decomposition of PCS adapted to the Decision Aid and a method for their analysis. The aim of the present analysis is to determine decision centers in order to build decision aid systems (DAS) adapted to the system controlled by decision makers. The conception of such DAS must use decision aid procedures and Artificial Intelligence techniques. This is consisting particularly in formalizing the Decision Aid Problem by using predicates, such procedures making the use of computer possible.