Towards better making a decision in speaker verification

Speaker verification is a process that accepts or rejects the identity claim of a speaker. How to make a decision is a critical problem; a threshold for decision-making critically determines performance of a speaker verification system. Traditional threshold estimation methods take only information conveyed by training data into consideration and, to a great extent, do not relate it to production data. It turns out that a speaker verification system with such threshold estimation suffers from poor performance in reality due to mismatches. In this paper, we propose several methods towards better decision-making in a practical speaker verification system. Our methods include the use of additional reliable statistical information for threshold estimation, elimination of abnormal data for better estimation of underlying statistics, and on-line incremental threshold update. To evaluate the performance of our methods, we have done simulations based on a baseline system, Gaussian Mixture Model, in both text-dependent and text-independent modes. Comparative results show that in contrast to the recent threshold estimation methods our methods yield considerably better performance, especially on miscellaneous mismatch conditions, in terms of generalization. Thus our methods provide a promising way for real speaker verification applications.     

软件可靠性评测的多因素决策模型

提出了一种利用软件生命周期过程中影响软件可靠性的软件质量特性进行软件可靠性评测的方法.从软件失效的机理出发,分析并提取了刻画软件可靠性各个方面的因素,同时借助模糊分析及专家系统的理论对软件可靠性因素进行了定量的划分和描述.针对软件可靠性评测的多因素决策问题,提出了一种基于随机森林的软件可靠性评测模型.通过蒙特卡罗模拟仿真建立了各个可靠性因素的概率模型以获得评测样本数据集,并在此数据集上对所提出的评测模型进行了评测和分析.实验结果表明,所提出的方法能够对软件可靠性进行准确的评测,且不依赖于特定的可靠性因素的先验概率,显著提高了软件可靠性的评测性能.同时验证了该方法能够克服小样本集上易出现的过拟合及表现力差的问题,具有较好的稳定性和鲁棒性.     

The role of project management in ineffective decision making within Agile software development projects

Decision making in traditional software development lies with the project manager. In contrast, Agile software development teams are empowered to make decisions, while the role of project manager has changed from one of command and control (i.e. to make decisions and ensure they are implemented) to one of a facilitator. This article argues that decision making in software development is not characterised by a sequence of isolated or exclusive decisions; rather, decisions are inter-related, with each decision leading to further decisions, the chain of which often spans the entire duration of a project. Over this extended period, there are several potential factors that can negatively affect the efficacy of decision making by Agile teams. One of the findings of this exploratory longitudinal study is that the high level of empowerment of a cohesive software development team undertaking an Agile project may be one of these negative factors, as empowered, cohesive teams can exhibit problems such as groupthink or the Abilene Paradox. This article therefore argues that the role of project manager in Agile development initiatives needs to be reassessed, with project managers taking on the role of devil's advocate in the decision-making process.     

Group decision making with multi-attribute interval data

The aim of this paper is to present a group decision making methodology, in which the decision information, including the attribute values, attribute weights and weights of decision makers, is expressed in interval data. An extended TOPSIS technique is twice used in the proposed method, which is first used to determine the weights of decision makers, and second used to rank the preference order of alternatives. There is no aggregation of decision information in decision process, except that the ideal decisions as auxiliary decision tools are used in decision process. We give a comparison with another method for group decision making to show the technical advance of reported method. Additionally, we also give a real life application for supplier selection and a discussion to test the effectiveness and practical implications of the proposed method.     

支持共识决策过程的群排序集结

为了支持共识决策过程,引入最大共识排序概念,设计了基于共识排序树的群排序集结算法。该算法能够从排序数据中发现最大共识排序和需要进一步协商的冲突项目。应用模拟数据进行实验,结果表明了这种计算方法的有效性。     

Fuzzy compromise programming for Group decision making

A multicriteria technique named fuzzy compromise programming is combined with a methodology known as group decision making under fuzziness to come up with a new technique that supports decision making with multiple criteria and multiple participants (or experts). All criteria (qualitative and quantitative) are modeled by way of fuzzy sets, utilizing the fact that criteria values in most water resources problems are vague, imprecise and/or ill defined. The involvement of multiple experts in the decision process is achieved by incorporating each participant's perception of criteria weights, best and worst criteria values, relative degrees of risk acceptance, as well as other parameters into the problem. The proposed methodology is illustrated with a case study taken from the literature, combined with the input of four expert individuals with diverse backgrounds. After processing the input from the experts, a group compromise decision is formulated.     

Group RFM analysis as a novel framework to discover better customer consumption behavior

The RFM model provides an effective measure for customers’ consumption behavior analysis, where three variables, namely, consumption interval, frequency, and money amount are used to quantify a customer’s loyalty and contribution. Based on the RFM value, customers can be clustered into different groups and the group information is very useful in market decision making. However, most previous works completely left out important characteristics of purchased products, such as their prices and lifetimes, and apply the RFM measure on all of a customer’s purchased products. This renders the calculation of the RFM value unreasonable or insignificant for customer analysis. In this paper, we propose a new framework called GRFM (for group RFM) analysis to alleviate the problem. The new measure method takes into account the characteristics of the purchased items so that the calculated the RFM value for the customers are strongly related to their purchased items and can correctly reflect their actual consumption behavior. Moreover, GRFM employs a constrained clustering method PICC (for Purchased Items-Constrained Clustering) that could base on a cleverly designed purchase pattern table to adjust original purchase records to satisfy various clustering constraints as well as to decrease re-clustering time. The GRFM allows a customer to belong to different clusters, and thus to be associated with different loyalties and contributions with respect to different characteristics of purchased items. Finally, the clustering result of PICC contains extra information about the distribution status inside each cluster that could help the manager to decide when is most proper to launch a specific sales promotion campaign. Our experiments have confirmed the above observations and suggest that GRFM can play an important role in building a personalized purchasing management system and an inventory management system.     

基于动态双极值模糊软集的群决策方法

针对实际问题中双极值模糊软集随时间变化的影响,定义了动态双极值模糊软集等概念,讨论了相关运算及性质。根据时间权重符合对数增长模型得到权重确定公式。利用集成思想定义双极值模糊软集的运算并给出几何加权平均算子的计算公式,将动态双极值模糊软集集成为综合双极值模糊软集。利用水平软集算出各对象的机会值,得出最优决策。通过实例分析证明此决策方法的合理性与可行性。     

Requirements engineering: making the connection between the software developer and customer

Requirements engineering are one of the most crucial steps in software development process. Without a well-written requirements specification, developer's do not know what to build, user's do not know what to expect, and there is no way to validate that the created system actually meets the original needs of the user. Much of the emphasis in the recent attention for a software engineering discipline has centered on the formalization of software specifications and their flowdown to system design and verification. Undoubtedly, the incorporation of such sound, complete, and unambiguous traceability is vital to the success of any project. However, it has been our experience through years of work (on both sides) within the government and private sector military industrial establishment that many projects fail even before they reach the formal specification stage. That is because too often the developer does not truly understand or address the real requirements of the user and his environment.The purpose of this research and report is to investigate the key players and their roles along with the existing methods and obstacles in Requirements Elicitation. The article will concentrate on emphasizing key activities and methods for gathering this information, as well as offering new approaches and ideas for improving the transfer and record of this information. Our hope is that this article will become an informal policy reminder/guideline for engineers and project managers alike. The success of our products and systems are largely determined by our attention to the human dimensions of the requirements process. We hope this article will bring attention to this oft-neglected element in software development and encourage discussion about how to effectively address the issue.     

Group decision making problems in a linguistic and dynamic context

The aim of this paper is to present a new model of decision support system for group decision making problems based on a linguistic approach and dynamic sets of alternatives. The model incorporates a mechanism that allows to manage dynamic decision situations in which some information about the problem is not constant in time. We assume that the set of alternatives can change during the decision making process. The model is presented in a mobile and dynamic context where the experts’ preferences can be incomplete. The linguistic approach is used to represent both the experts’ preferences about the alternatives and the agreement degrees to manage the change of some alternatives. A prototype of such mobile decision support system in which the experts use mobile devices to provide their linguistic preferences at anytime and anywhere has been implemented. In such a way, we provide a new linguistic group decision making framework that is mobile and dynamic.     

基于FAHP的信息安全风险群组决策评估方法*

结合当前信息安全风险评估的特点和发展现状,提出一种结合FAHP和群组决策的风险量化评估方法。该方法利用FAHP处理主观评估判断结果,实现了综合考虑风险发生概率和风险损失的偏好排序;在评估过程中引入群组决策,建立群组偏好排序的线性规划模型,以降低个体评估决策的不确定性。以某政务公开信息系统脆弱性评估为例,验证了评估方法的合理性和可行性,结果表明该评估模型计算简单且排序稳定性好。     

Group decision making with linguistic preference relations with application to supplier selection

Linguistic preference relation is a useful tool for expressing preferences of decision makers in group decision making according to linguistic scales. But in the real decision problems, there usually exist interactive phenomena among the preference of decision makers, which makes it difficult to aggregate preference information by conventional additive aggregation operators. Thus, to approximate the human subjective preference evaluation process, it would be more suitable to apply non-additive measures tool without assuming additivity and independence. In this paper, based on λ-fuzzy measure, we consider dependence among subjective preference of decision makers to develop some new linguistic aggregation operators such as linguistic ordered geometric averaging operator and extended linguistic Choquet integral operator to aggregate the multiplicative linguistic preference relations and additive linguistic preference relations, respectively. Further, the procedure and algorithm of group decision making based on these new linguistic aggregation operators and linguistic preference relations are given. Finally, a supplier selection example is provided to illustrate the developed approaches.     

Group decision making with incomplete fuzzy linguistic preference relations

The aim of this paper is to propose a procedure to estimate missing preference values when dealing with incomplete fuzzy linguistic preference relations assessed using a two‐tuple fuzzy linguistic approach. This procedure attempts to estimate the missing information in an individual incomplete fuzzy linguistic preference relation using only the preference values provided by the respective expert. It is guided by the additive consistency property to maintain experts' consistency levels. Additionally, we present a selection process of alternatives in group decision making with incomplete fuzzy linguistic preference relations and analyze the use of our estimation procedure in the decision process. © 2008 Wiley Periodicals, Inc.     

基于前景理论的群体灰靶决策方法

考虑决策者关于各指标均有期望灰靶对群体决策的影响,提出一种基于前景理论的群体灰靶决策方法.该方法以期望灰靶为参考点来定义前景价值函数,利用奖优罚劣的线性变换算子对前景价值进行规范化处理,能够充分反映评价值是否中靶.根据群体意见一致性以及最高和最低评价对决策结果偏差的影响,构建决策者权重确定模型,并根据综合前景值的正负判断方案是否中靶.最后通过突发事件应急预案选择问题说明了该模型的可行性和有效性.     

加权犹豫模糊集的群决策方法

对于犹豫模糊元中的不同隶属度值赋予不同的权重,由此构造出一种应用范围更广、更符合实际需要的犹豫模糊集合 ----- 加权犹豫模糊集合.针对加权犹豫模糊集中的加权犹豫模糊元,定义了加权犹豫模糊集合和加权犹豫模糊元的并、交、余、数乘和幂等运算及其运算法则,并讨论它们的运算性质;同时,给出加权犹豫模糊元的得分函数和离散函数,进而给出一种比较加权犹豫模糊元的排序法则.在此基础上,提出两类集成算子:加权犹豫模糊元的加权算术平均算子和加权犹豫模糊元的加权几何平均算子,并针对专家权重(已知和未知)的两种情形,将加权犹豫模糊集合应用于群决策,给出两种基于加权犹豫模糊集合的群决策方法.最后,通过一个应用实例表明所提出的群决策方法的有效性和实用性.     

Integrating simulation modelling and multi criteria decision making for customer focused scheduling in job shops

Today, customer centricity is an important strategy in business-to-business markets and manufacturing companies need decision support systems that provide adequate information for customer centric applications. This study proposes an integrated decision support system that combines simulation modelling and multi-criteria decision making. More specifically, job shop lot streaming problem is dealt with, and it is aimed to determine the best dispatching rules to schedule batches on machines. To this aim, three renowned performance-oriented criteria; (i) mean flow time, (ii) percentage of tardy orders, (iii) makespan and one customer-oriented criterion; (iv) mean percentage deviation from the customer expectations are considered. Effect of different classical and customer-oriented dispatching rules on these performance criteria are investigated. The performance criteria are weighted using analytical hierarchy process by considering the level of bottleneck resource utilization and customer importance weights. The results reveal that customer-oriented dispatching rules provide better outcomes in case of high level of bottleneck resource utilization and high fluctuation amongst the customer importance weights.     

Group decision making procedure considering preference strength under incomplete information

This article deals with the multiple criteria decision making problem with incomplete information when multiple decision makers (Multiple Criteria Group Decision Making: MCGDM) are involved. Usually decision makers (DMs) are willing or able to provide only incomplete information, because of time pressure, lack of knowledge or data, and their limited expertise related to the problem domain. There have only been a few studies considering incomplete information in group settings. We also consider the case where importance weights are given incompletely. This article suggests the possibility that individually optimized results can be used to build group consensus. Individual optimization results by pairwise dominance, contain useful information in forming consensus, such as, ordinal rankings or preference intensity of an alternative over the others. Rather than using ordinal rankings for aggregation which do not consider preference strength, we suggest a procedure that takes account of individual DMs' preference strength.     

A decision model for managing software development projects

The paper first examines some issues that hinder the effective management of, and decision-making on, quality software development process and products delivery by practitioners. It then generates a decision model for managing software development projects. The model uses four concepts: mappability, accountability, interoperability and controllability in decision-making which is assumed to be based on a set of indicators that link task status of the development process and its quality assessment to the responsible authorities. The quality of the tasks, and hence, of the deliverables is measured using four attributes: completeness, correctness, consistency and compliance. A web-based example implementation is then discussed. We thus show that the model is flexible, extensible and scalable. Implementation challenges and implications are then discussed.