Group Decision and Negotiation最新文献

In the era of big data, information leakage during medical consultation has become a crucial factor in patients’ decision-making. This paper presents an intelligent medical decision model that considers patient privacy. The model utilizes data synthesized through a generative adversarial network (GAN) for intelligent training, ensuring privacy protection. First, we formulate a risk-based decision model for three different alternative medical consultation modes, analyzing decision rules related to visiting distance and disease probability. Next, we construct a data-driven intelligent medical decision framework. To address privacy concerns, we employ GAN to generate synthetic data from historical patient records, which is seamlessly incorporated into the decision framework to derive decision rules. Finally, specific patient data is utilized to make informed medical decisions. We validated our model using the random forest algorithm and liver disease patients’ medical decisions. Empirical findings demonstrate that the GAN-based synthetic data improves the nearest-neighbor distance ratio by 12.4% compared to synthetic data with Gaussian noise, thereby enhancing data privacy. Additionally, the GAN-based prediction models outperform the models trained on real data, achieving a notable increase of 6.3% and 4.1% in average accuracy and F1 score, respectively. Notably, the GAN-based intelligent decision-making models surpass four other baseline medical visit decision-making methods with an impressive accuracy of 74.0%. In conclusion, our proposed intelligent medical decision-making model effectively prioritizes user data privacy while enhancing the quality of medical decision-making.

Requirements elicitation and analysis tasks in user-centered design (UCD) are pivotal for assessing digital systems’ quality and costs. However, these tasks often face challenges due to limited user involvement. This stems from unclear guidelines on how to conduct activities and engage users effectively to achieve their goals during the development process. This study explored how the integration of collaboration engineering (CE) principles with UCD approach could address these challenges. Using an Applied Science / Engineering approach, a UCD-CE process was designed drawing on the Six-layer model of Collaboration. This model aligns the CE steps with UCD principles (why), practices (what), and methods (how). Data collection tools included structured interviews, questionnaires, and observations, supported by techniques like user stories and dialogues, as well as thinkLets, and patterns of collaboration. Formative and summative evaluations were used to validate the UCD-CE process; and the results underscore its strengths, particularly its efficiency in helping users to complete tasks on time, reducing effort in reaching common goals, fostering high user satisfaction, promoting creativity and productivity, ensuring ease-of-use and learnability, and delivering comprehensive outcomes in requirements elicitation and analysis tasks during the development process. Future research aims to assess the practicality of UCD-CE integration in reinforcing user involvement during the UCD design phase.

Psychological safety (PS), the feeling of being comfortable to express one’s ideas or opinions in teams, is a key determinant of successful global virtual teams (GVT). Even though considerable knowledge exists about its antecedents, it is unknown how team-based interventions (TBI) and technology-based interventions (digital reminder nudges, DRN) foster PS among team members. Based on a survey involving 235 participants, our data show that TBI and DRN foster psychological safety in GVT. However, only the effect of TBI on psychological safety can be explained with a higher-quality coordination process. It remains unclear what causal mechanism explains the effect of DRN. These findings contribute to the literature on PS by showing that TBI facilitate effective coordination processes and to the literature on digital nudges by demonstrating that technology-based reminders drive PS.

In group decisions, achieving consensus is important, because it increases commitment to the result. For cooperative groups, Combinatorial Multicriteria Acceptability Analysis (CMAA) is a group decision framework that can achieve consensus efficiently. It is based on a novel Combinatorial Acceptability Entropy (CAE) consensus metric. As an output measure, the CAE metric is unique in its ability to identify the evaluations that have the greatest impact on consensus and to prevent premature consensus. This paper is intended to complement the original CMAA publication by providing additional insights into the CAE consensus metric. The design requirements for the CAE algorithm are presented, and it is shown how these requirements follow from the properties of cooperative decisions. The CAE-based consensus-building algorithm is contrasted both qualitatively and quantitatively with a representative example of the conventional input distance and input averaging approach to multi-criteria consensus-building. A simulation experiment illustrates the ability of the CAE-based algorithm to converge quickly to the correct decision as defined for cooperative decisions. The metric is able to meet a new, more stringent definition of hard consensus. The CAE approach highlights the need to distinguish between competitive and cooperative group decisions. Attention in the literature has been paid almost exclusively to the former type; the CAE approach demonstrates the greater efficiency and effectiveness that can be achieved with an approach that is designed specifically for the latter.

Macro-task crowdsourcing presents a promising approach to address wicked problems like climate change by leveraging the collective efforts of a diverse crowd. Such macro-task crowdsourcing requires facilitation. However, in the facilitation process, traditionally aggregating and synthesizing text contributions from the crowd is labor-intensive, demanding expertise and time from facilitators. Recent advancements in large language models (LLMs) have demonstrated human-level performance in natural language processing. This paper proposes an abstract design for an information system, developed through four iterations of a prototype, to support the synthesis process of contributions using LLM-based natural language processing. The prototype demonstrated promising results, enhancing efficiency and effectiveness in synthesis activities for macro-task crowdsourcing facilitation. By streamlining the synthesis process, the proposed system significantly reduces the effort to synthesize content, allowing for stronger integration of synthesized content into the discussions to reach consensus, ideally leading to more meaningful outcomes.

We propose an alternative decision-making methodology based on adopting a mixed risk-averse and risk-taking behavior, improving the objectivity of decision-making. We demonstrate the methodology by prioritizing Iranian tourism centers’ activity under pandemic conditions, providing insights to policymakers on those to keep active or reduce the activity of – hence, those worth developing ahead of future disease outbreaks. This research follows a three-step methodology. First, criteria for evaluation are identified and categorized into tourist attractions, infrastructure, and healthcare dimensions. Second, criterion weights are calculated based on expert opinions, collected using a best-worst method-based questionnaire. Third, tourism centers are evaluated by employing risk-averse and risk-taking best-worst methods. We identify popular attractions, general services, and drugstore accessibility as the primary indicators of tourist attractions, infrastructure, and healthcare, respectively. By clustering tourism centers using K-means algorithm, we find that, in order, the cities of Semnan, Kerman and Zahedan are the tourism centers most suited to staying active during disease outbreaks. For multi-criteria decision-making problems that rely on experts’ evaluations, the proposed methodology can improve the reliability of decision-making. The methodology and framework presented can be used to support various types of decision-making, including evaluation, ranking, selection or sorting.

In Large-Scale Group Decision-Making (LSGDM), effectively implementing consensus models is pivotal for managing decision complexity. While trust-based LSGDM has garnered attention, there remains a need for deeper insights into the dynamics of interexpert trust and the impact of authority effects on the decision-making process. This study introduces a sophisticated model for large-scale group decision-making, incorporating considerations of expert “trustworthiness-authority.” Initially, the study assesses the trustworthiness of experts based on social network relationships and opinion similarity while using background information and consensus levels to establish their authority. Subsequently, experts are categorized into four distinct regions based on their trustworthiness and authority assessments. Furthermore, tailored consensus adjustment methods are proposed for each region based on social contagion theory to facilitate consensus achievement. Additionally, a case study is conducted to demonstrate the rationality and effectiveness of the proposed LSGDM model, considering expert “trustworthiness-authority.” Finally, the necessity and superiority of the proposed model are further verified through comparison analysis and sensitivity analysis.

As a process for ensuring the agreeable degree of individual opinions, consensus analysis is crucial for GDM. This paper focuses on the adaptive consensus mechanism. That's, different adjustment strategies are employed for various consensus levels. Unlike the feedback iteration method, this paper introduces an optimization model-based consensus-reaching procedure. To do this, optimal models are built to determine the minimum consensus adjustment at different levels. Then, the individual minimum consensus adjustment is analyzed, and the inconsistency between individual and group minimum consensus adjustments is concluded. After that, consensus adjustment cooperative games at three levels are proposed to allocate the total minimum consensus adjustment in view of the comprehensive evaluation. We can obtain the coalitional stability allocation scheme using the core of constructed cooperative games. Additionally, core-Nash bargaining games at three levels are proposed to ensure the fairness and coalitional stability of allocation results. Finally, a numerical example is offered to indicate the application of the new theoretical developments.

Learning an opponent’s preferences in bilateral multi-issue automated negotiations can lead to more favorable outcomes. However, existing opponent models can fail in negotiation contexts when their assumptions about opponent behaviors differ from actual behavior patterns. Although integrating broader behavioral assumptions into these models could be beneficial, it poses a challenge because the models are designed with specific assumptions. Therefore, this study proposes an adaptable opponent model that integrates a general behavioral assumption. Specifically, the proposed model uses Bayesian learning (BL), which can apply various behavioral assumptions by considering the opponent’s entire bidding sequence. However, this BL model is computationally infeasible for multi-issue negotiations. Hence, current BL models often impose constraints on their hypothesis space, but these constraints about the utility function’s shape significantly sacrifice accuracy. This study presents a novel scalable BL model that relaxes these constraints to improve accuracy while maintaining linear time complexity by separately learning each parameter of a utility function. Furthermore, we introduce a general assumption that the opponent’s bidding strategy follows a concession-based pattern to enhance adaptability to various negotiation contexts. We explore three likelihood function options to implement this assumption effectively. By incorporating these options into the proposed scalable model, we develop three scalable concession-driven opponent models using Bayesian learning (COMB). Experiments across 45 negotiation domains using 15 basic agents and 15 finalists from the automated negotiating agents competition demonstrate the proposed scalable model’s higher accuracy than existing scalable models. COMB models show higher adaptability to various negotiation contexts than state-of-the-art models.

This work proposes and studies a dynamic model of two bargaining parties exchanging offers over time, considering their confidence about the share of the “pie” they obtain, which translates into expectations regarding the outcome of the bargaining process. The model predicts the sequence of offers as well as the final agreement for given confidence parameters. A mathematical analysis of the model shows the outcome is an Asymmetric Nash Bargaining Solution with exponents determined by the bargainers’ confidence. Moreover, a compensation effect can be found between confidence and risk aversion. This work also considers that confidence levels of bargainers might change during the negotiation, and we conduct a comprehensive simulation study to analyze the effect of such changes. Through Monte-Carlo simulation, we show that a bargainer is better off if its confidence increases, but the advantage is lost if the other party’s confidence increases in a similar way. In that case, concessions are smaller and negotiations last longer. Changing confidence parameters make the outcome harder to predict, as it will depend more on the final confidence than the initial one. The simulations also show that the average size of concessions, and therefore the final agreement, depend not only on whether confidence increases or decreases, but also on the change rate, with stronger effects observed when change accelerates towards the end of the process.