Search results

There is an increasing interest in the use of knowledge graphs to represent real-world knowledge and a common need to manage imprecise knowledge in many real-world applications. This paper aims to study approaches to solve flexible queries over knowledge graphs.

By introducing fuzzy logic in the query answering process, the authors are able to obtain a novel algorithm to solve flexible queries over knowledge graphs. This approach is implemented in the FUzzy Knowledge Graphs system, a software tool with an intuitive user-graphical interface.

This approach makes it possible to reuse semantic web standards (RDF, SPARQL and OWL 2) and builds a fuzzy layer on top of them. The application to a use case shows that the system can aggregate information in different ways by selecting different fusion operators and adapting to different user needs.

This approach is more general than similar previous works in the literature and provides a specific way to represent the flexible restrictions (using fuzzy OWL 2 datatypes).

Subway systems are highly susceptible to external disturbances from emergencies, triggering a series of consequences such as the paralysis of the internal network transportation functions, causing significant economic and safety losses to cities. Therefore, it is necessary to analyze the factors affecting the resilience of the subway system to reduce the impact of disaster incidents.

Using the interval type-2 fuzzy linguistic term set and the K-medoids clustering algorithm, this paper improves the Decision-Making Trial and Evaluation Laboratory (DEMATEL) method to construct a subway resilience factor analysis model for emergencies. Through comparative analysis, this study confirms the superior performance of the proposed approach in enhancing the precision of the DEMATEL method.

The results indicate that the operation and management level of emergency command organizations is the key resilience factors of subway operations in China. Furthermore, based on real case analyses, the corresponding suggestions and measures are put forward to improve the overall operation resilience level of the subway.

This paper identifies four emergency scenarios and 15 resilience factors affecting subway operations through literature review and expert consultation. The improved fuzzy DEMATEL method is applied to explore the levels of influence and causal mechanisms among the resilience factors of the subway system under the four emergency scenarios.

The purpose of this study is to build a consensus model of social network group decision-making (SNGDM) based on improved PageRank algorithm. By objectively and fairly measuring the evaluation ability of participants in the decision-making process, the authors can improve the fairness and authenticity of the weight solution of decision-makers (DM) in the decision-making process. This ensures the reliability of the final group consensus results.

This study mainly includes six parts: preference expression, calculation of DM's weight, preference aggregation, consensus measurement, opinion adjustment and alternative selection. First, Pythagorean fuzzy expression is introduced to express the preference of DMs, which expands the scope of preference expression of DMs. Second, based on the social network structure among DMs, the process of “mutual judgment” among DMs is increased to measure the evaluation ability of DMs. On this basis, the PageRank algorithm is improved to calculate the weight of DMs. This makes the process of reaching consensus more objective and fair. Third, in order to minimize the evaluation difference between groups and individuals, a preference aggregation model based on plant growth simulation algorithm (PGSA) is proposed to aggregate group preferences. Fourth, the consensus index of DMs is calculated from three levels to judge whether the consensus degree reaches the preset value. Fifth, considering the interaction of DMs in the social network, the evaluation value to achieve the required consensus degree is adjusted according to the DeGroot model to obtain the overall consensus. Finally, taking the group preference as the reference, the ranking of alternatives is determined by using the Pythagorean fuzzy score function.

This paper proposes a consensus model of SNGDM based on improved PageRank algorithm to aggregate expert preference information. A numerical case of product evaluation is introduced, and the feasibility and effectiveness of the model are explained through sensitivity analysis and comparative analysis. The results show that this method can solve the problem of reaching consensus in SNGDM.

Different DMs may have different judgment criteria for the same decision-making problem, and the angle and depth of considering the problem will also be different. By increasing the process of mutual evaluation of DMs, the evaluation ability of each DM is judged only from the decision-making problem itself. In this way, the evaluation opinions recognized by most DMs will form the mainstream of opinions, and the influence of corresponding DMs will increase. Therefore, in order to improve the fairness and reliability of the consensus process, this study measures the real evaluation ability of DMs by increasing the “mutual judgment” process. On this basis, the defect of equal treatment of PageRank algorithm in calculating the weight of DMs is improved. This ensures the authenticity and objectivity of the weight of DMs. That is to improve the effectiveness of the whole evaluation mechanism. This method considers both the influence of DMs in the social network and their own evaluation level. The weight of DMs is calculated from two aspects: sociality and professionalism. It provides a new method and perspective for the calculation of DM’s weight in SNGDM.

User feedback inferred from the user's search-time behavior could improve the learning to rank (L2R) algorithms. Click models (CMs) present probabilistic frameworks for describing and predicting the user's clicks during search sessions. Most of these CMs are based on common assumptions such as Attractiveness, Examination and User Satisfaction. CMs usually consider the Attractiveness and Examination as pre- and post-estimators of the actual relevance. They also assume that User Satisfaction is a function of the actual relevance. This paper extends the authors' previous work by building a reinforcement learning (RL) model to predict the relevance. The Attractiveness, Examination and User Satisfaction are estimated using a limited number of the features of the utilized benchmark data set and then they are incorporated in the construction of an RL agent. The proposed RL model learns to predict the relevance label of documents with respect to a given query more effectively than the baseline RL models for those data sets.

In this paper, User Satisfaction is used as an indication of the relevance level of a query to a document. User Satisfaction itself is estimated through Attractiveness and Examination, and in turn, Attractiveness and Examination are calculated by the random forest algorithm. In this process, only a small subset of top information retrieval (IR) features are used, which are selected based on their mean average precision and normalized discounted cumulative gain values. Based on the authors' observations, the multiplication of the Attractiveness and Examination values of a given query–document pair closely approximates the User Satisfaction and hence the relevance level. Besides, an RL model is designed in such a way that the current state of the RL agent is determined by discretization of the estimated Attractiveness and Examination values. In this way, each query–document pair would be mapped into a specific state based on its Attractiveness and Examination values. Then, based on the reward function, the RL agent would try to choose an action (relevance label) which maximizes the received reward in its current state. Using temporal difference (TD) learning algorithms, such as Q-learning and SARSA, the learning agent gradually learns to identify an appropriate relevance label in each state. The reward that is used in the RL agent is proportional to the difference between the User Satisfaction and the selected action.

Experimental results on MSLR-WEB10K and WCL2R benchmark data sets demonstrate that the proposed algorithm, named as SeaRank, outperforms baseline algorithms. Improvement is more noticeable in top-ranked results, which usually receive more attention from users.

This research provides a mapping from IR features to the CM features and thereafter utilizes these newly generated features to build an RL model. This RL model is proposed with the definition of the states, actions and reward function. By applying TD learning algorithms, such as the Q-learning and SARSA, within several learning episodes, the RL agent would be able to learn how to choose the most appropriate relevance label for a given pair of query–document.

A successful process of design concept evaluation has positive influence on subsequent processes. This study aims to consider the evaluation information at multiple stages and the interaction among evaluators and improve the credibility of evaluation results.

This paper proposes a multi-stage approach for design concept evaluation based on complex network and bounded confidence. First, a network is constructed according to the evaluation data. Depending on the consensus degree of evaluation opinions, the number of evaluation rounds is determined. Then, bounded confidence rules are applied for the modification of preference information. Last, a planning function is constructed to calculate the weight of each stage and aggregate information at multiple evaluation stages.

The results indicate that the opinions of the evaluators tend to be consistent after multiple stages of interactive adjustment, and the ordering of design concept alternatives tends to be stable with the progress of the evaluation.

Updating preferences according to the bounded confidence rules, only the opinions within the trust threshold are considered. The attribute information of the node itself is inadequately considered.

This method addresses the need for considering the evaluation information at each stage and minimizes the impact of disagreements within the evaluation group on the evaluation results.

The main goal of this paper is to develop novel Tω(weakest t-norm)-based fuzzy arithmetic operations to analyze the intuitionistic fuzzy reliability of Printed Circuit Board Assembly (PCBA) using fault tree.

The paper proposes a fuzzy fault tree analysis (FFTA) method for evaluating the intuitionistic fuzzy reliability of any nonrepairable system with uncertain information about failures of system components. This method uses a fault tree for modeling the failure phenomenon of the system, triangular intuitionistic fuzzy numbers (TIFNs) to determine data uncertainty, while novel arithmetic operations are adopted to determine the intuitionistic fuzzy reliability of a system under consideration. The proposed arithmetic operations employ Tω(weakest t-norm) to minimize the accumulating phenomenon of fuzziness, whereas the weighted arithmetic mean is employed to determine the membership as well as nonmembership degrees of the intuitionistic fuzzy failure possibility of the nonrepairable system. The usefulness of the proposed method has been illustrated by inspecting the intuitionistic fuzzy failure possibility of the PCBA and comparing the results with five other existing FFTA methods.

The results show that the proposed FFTA method effectively reduces the accumulating phenomenon of fuzziness and provides optimized degrees of membership and nonmembership for computed intuitionistic fuzzy reliability of a nonrepairable system.

The paper introduces Tω(weakest t-norm) and weighted arithmetic mean based operations for evaluating the intuitionistic fuzzy failure possibility of any nonrepairable system in an uncertain environment using a fault tree.

The development of social networks enhances the interaction between people, which brings new challenges to the research of group decision-making (GDM). This study aims at the problem that the synergy and redundancy due to interaction among decision-makers are ignored in the previous GDM, a trust-enhanced consensus reaching model based on interaction among decision-makers with incomplete preferences is proposed.

Firstly, confidence level is introduced to improve the hesitation phenomenon that should be considered when calculating trust degree; Secondly, a new trust propagation operator is developed to deal with indirect trust relationships; Thirdly, trust degree is transformed into interaction index to quantify the synergy and redundancy in decision-making. Fuzzy capacities of decision-makers are used to replace traditional weights, and the final scores of alternatives are obtained through Choquet integral.

The proposed model using fuzzy capacity can reflect the synergy or redundancy among decision-makers and improve the accuracy of final ranking result and reduce the loss of information.

This study proposes a trust-enhanced consensus reaching model, which develops a new trust propagation operator to ensure the continuous attenuation of trust in propagation process. And the proposed model uses fuzzy capacity to improve the enhancement or attenuation on the scores of alternatives.

This study attempts to identify and explicate the unique segmentation of the increasingly growing virtual reality (VR) user market based on the user experience. Consequently, it collects five hundred forty-five online survey questionnaires through the Prolific platform and deploys cluster analysis to identify mutually exclusive groups of VR users. The research variable, user experience, contains 16 indicators explained by four dimensions. As a result, this study is able to unveil three mutually exclusive markets which are labeled as (1) beginner, (2) aficionado, and (3) utilitarian. The unique features of these three groups are further compared based on their VR tour behaviors. In the conclusion section, it offers managerial implications for devising novel marketing strategies.

This paper aims to examine and compare extant literature on the application of multi-criteria decision-making (MCDM) techniques in humanitarian operations (HOs) and humanitarian supply chains (HSCs). It identifies the status of existing research in the field and suggests a roadmap for academicians to undertake further research in HOs and HSCs using MCDM techniques.

The paper systematically reviews the research on MCDM applications in HO and HSC domains from 2011 to 2022, as the field gained traction post-2004 Indian Ocean Tsunami phenomena. In the first step, an exhaustive search for journal articles is conducted using 48 keyword searches. To ensure quality, only those articles published in journals featuring in the first quartile of the Scimago Journal Ranking were selected. A total of 103 peer-reviewed articles were selected for the review and then segregated into different categories for analysis.

The paper highlights insufficient high-quality research in HOs that utilizes MCDM methods. It proposes a roadmap for scholars to enhance the research outcomes by advocating adopting mixed methods. The analysis of various studies revealed a notable absence of contextual reference. A contextual mind map specific to HOs has been developed to assist future research endeavors. This resource can guide researchers in determining the appropriate contextual framework for their studies.

This paper will help practitioners understand the research carried out in the field. The aspiring researchers will identify the gap in the extant research and work on future research directions.

To the best of the authors’ knowledge, this is the first literature review on applying MCDM in HOs and HSCs. It summarises the current status and proposes future research directions.

Complex cost structures and multiple conflicting objectives make selecting an appropriate cloud service difficult. The purpose of this study is to propose a novel group consensus decision making method for cloud services selection with knowledge deficit by trust functions.

This article proposes a knowledge deficit-based multi-criteria group decision-making (MCGDM) method for cloud-service selection based on trust functions. Firstly, the concept of trust functions and a ranking method is developed to express the decision-making opinions. Secondly, a novel 3D normalized trust degree (NTD) is defined to measure the consensus levels. Thirdly, a knowledge deficit-based interactive consensus model is proposed for the inconsistent experts to modify their decision opinions. Finally, a real case study has been carried out to illustrate the framework and compare it with other methods.

The proposed method is practical and effective which is verified by the real case study. Knowledge deficit is an important concept in cloud service selection which is verified by the comparison of the proposed recommended mechanism based on KDD with the conventional recommended mechanism based on average value. A 3D NTD which considers three values (trust, not trust and knowledge deficit) is defined to measure the consensus levels. A knowledge deficit-based interactive consensus model is proposed to help decision-makers reach group consensus. The proposed group consensus model enables the inconsistent decision-makers to accept the revised opinions of those with less knowledge deficit, rather than accepting the recommended opinions averagely.

The proposed a knowledge deficit-based MCGDM cloud service selection method considers group consensus in cloud service selection. The concept of knowledge deficit is considered in modeling the group consensus measuring and reaching method.