Consensus Information Research Articles

Tensor subspace clustering using consensus tensor low-rank representation

Recently, subspace clustering (SC) has received an increasing amount of attention. Generally, SC methods are implemented on sample data in matrix form. Even for multiway or tensor data, which is prevalent in reality, traditional SC methods should convert each sample to a vector and then form a data matrix. However, the vectorization process will damage the underlying inherent spatial structure of the data. To address this problem, in this paper, we propose a novel consensus tensor low-rank representation (CTLRR) method, which is directly implemented on tensor data. First, the tensor nuclear norm (TNN) and t-product, which is defined as the multiplication of two tensors, are employed to model the data and obtain the low-rank representation tensor. Second, spectral clustering is unified into CTLRR to explore the consensus information among the low-rank representation tensor, which helps obtain the final fusion similarity matrix with the k connected components, where k is the number of clusters. Third, the cluster structure is characterized, and the clustering performance is substantially improved. Last, an optimization procedure for CTLRR is presented. Experimental results on certain challenging datasets show that the proposed CTLRR method outperforms the state-of-the-art methods.

Beyond Shared Subspace: A View-Specific Fusion for Multi-View Multi-Label Learning

In multi-view multi-label learning (MVML), each instance is described by several heterogeneous feature representations and associated with multiple valid labels simultaneously. Although diverse MVML methods have been proposed over the last decade, most previous studies focus on leveraging the shared subspace across different views to represent the multi-view consensus information, while it is still an open issue whether such shared subspace representation is necessary when formulating the desired MVML model. In this paper, we propose a DeepGCN based View-Specific MVML method (D-VSM) which can bypass seeking for the shared subspace representation, and instead directly encoding the feature representation of each individual view through the deep GCN to couple with the information derived from the other views. Specifically, we first construct all instances under different feature representations into the corresponding feature graphs respectively, and then integrate them into a unified graph by integrating the different feature representations of each instance. Afterwards, the graph attention mechanism is adopted to aggregate and update all nodes on the unified graph to form structural representation for each instance, where both intra-view correlations and inter-view alignments have been jointly encoded to discover the underlying semantic relations. Finally, we derive a label confidence score for each instance by averaging the label confidence of its different feature representations with the multi-label soft margin loss. Extensive experiments have demonstrated that our proposed method significantly outperforms state-of-the-art methods.

Hypnotics and infections: disproportionality analysis of the U.S. Food & Drug Administration adverse event reporting system database.

There is no consensus information on infections associated with nonbenzodiazepines. Knowledge about infections related to newly marketed hypnotics (orexin receptor antagonists and melatonin receptor agonists) is scarce. The study aimed to detect infection signals for nonbenzodiazepines, orexin receptor antagonists, and melatonin receptor agonists by analyzing data from the U.S. Food & Drug Administration adverse event reporting system. A disproportionality analysis was performed to quantitatively detect infection signals for hypnotics by calculating the reporting odds ratio and the 95% confidence interval. Data registered in the U.S. Food & Drug Administration adverse event reporting system from 2010-2020 were retrieved. A total of 3,092 patients with infection were extracted for the 3 classes of hypnotic drugs. Nonbenzodiazepines were associated with a higher disproportionality of infections (reporting odds ratio: 1.10; 95% confidence interval, 1.06-1.14). The association of infections was not present for melatonin receptor agonists (reporting odds ratio: 0.86; 95% confidence interval, 0.74-1.00) and orexin receptor antagonists (reporting odds ratio: 0.19; 95% confidence interval, 0.15-0.25). Significant reporting associations were identified for nonbenzodiazepines concerning the categories of bone and joint infections, dental and oral soft tissue infections, upper respiratory tract infections, and urinary tract infections. Nonbenzodiazepines had a positive signal for infections, while orexin receptor antagonists and melatonin receptor agonists had a negative signal. More research needs to be conducted to confirm this relationship. Meng L, Huang J, He Q, etal. Hypnotics and infections: disproportionality analysis of the U.S. Food & Drug Administration adverse event reporting system database. J Clin Sleep Med. 2022;18(9):2229-2235.

Multi-layer manifold learning for deep non-negative matrix factorization-based multi-view clustering

Multi-view data clustering based on Non-negative Matrix Factorization (NMF) has been commonly used for pattern recognition by grouping multi-view high-dimensional data by projecting it to a lower-order dimensional space. However, the NMF framework fails to learn the accurate lower-order representation of the input data if it exhibits complex and non-linear relationships. This paper proposes a deep non-negative matrix factorization-based framework for effective multi-view data clustering by uncovering both the non-linear relationships and the intrinsic components of the data. Both the consensus and complementary information present in multiple views are sufficiently learned in the proposed framework with the effective use of constraints such as normalized cut-type and orthogonal. The optimal manifold of multi-view data is effectively incorporated in all layers of the framework. Extensive experimental results show the proposed method outperforms state-of-the-art multi-view matrix factorization-based methods.

Orthogonal multi-view tensor-based learning for clustering

Multi-view spectral clustering aims to improve the performance of spectral clustering through multi-view data. Many multi-view spectral clustering methods have been proposed recently and achieved promising performance. Among these methods, most of them are designed to pursue numerical consistency in multi-view similarity matrices. However, each similarity matrix has its unique statistic distribution, which makes it not appropriate to seek numerical consistency in multi-view similarity matrices or directly average the multi-view similarity matrices. To overcome the aforementioned problem, we propose a novel Orthogonal Multi-view Tensor-based Learning for clustering, abbreviated as OMTL. Specifically, OMTL introduces an orthogonal matrix factorization to eliminate the view-specific statistic distribution and preserve the intrinsic clustering structure of each view, which fully considers the consensus information contained in multiple views to boost multi-view spectral clustering performance. Further, we employ a low-rank tensor constraint to explore the high order correlations among multiple views. By designing an alternating direction method of multipliers (ADMM) based optimization algorithm, the intrinsic similarity matrix of multi-view data can be efficiently learned for spectral clustering. Extensive experiments on several benchmark datasets have illustrated the superior clustering performance of the proposed method compared to several state-of-the-art multi-view clustering methods.

Contrastive and attentive graph learning for multi-view clustering

Graph-based multi-view clustering aims to take advantage of multiple view graph information to provide clustering solutions. The consistency constraint of multiple views is the key of multi-view graph clustering. Most existing studies generate fusion graphs and constrain multi-view consistency by clustering loss. We argue that local pair-view consistency can achieve fine-modeling of consensus information in multiple views. Towards this end, we propose a novel Contrastive and Attentive Graph Learning framework for multi-view clustering (CAGL). Specifically, we design a contrastive fine-modeling in multi-view graph learning using maximizing the similarity of pair-view to guarantee the consistency of multiple views. Meanwhile, an Att-weighted refined fusion graph module based on attention networks to capture the capacity difference of different views dynamically and further facilitate the mutual reinforcement of single view and fusion view. Besides, our CAGL can learn a specialized representation for clustering via a self-training clustering module. Finally, we develop a joint optimization objective to balance every module and iteratively optimize the proposed CAGL in the framework of graph encoder–decoder. Experimental results on six benchmarks across different modalities and sizes demonstrate that our CAGL outperforms state-of-the-art baselines.

Large-Scale Multi-View Clustering via Fast Essential Subspace Representation Learning

Large-scale Multi-View Clustering (LMVC) is a hot research problem in the fields of signal processing and machine learning, and many anchor-based multi-view subspace clustering algorithms are proposed in recent years. However, most existing methods usually concentrate on the issue of reducing the time cost and ignore the exploration of the complementary information during the clustering process. To this end, we propose a Fast Essential Subspace Representation Learning (FESRL) method for large-scale multi-view subspace clustering. Specifically, FESRL introduces the orthogonal transformation to investigate both the complementary and consensus information across multiple views. The essential subspace representation can be learned in a linear time cost. Experiments conducted on several benchmark datasets illustrate the competitiveness of the proposed method.

Three-way group consensus decision based on hierarchical social network consisting of decision makers and participants

In group decision making (GDM) of organization management, information interaction usually occurs via the hierarchical social networks. In the hierarchical social networks, the relation and opinion exchange between decision makers and participants can influence the consensus reaching of three-way group decision. In this paper, we deeply explore the three-way group decision consensus reaching by constructing adjustment mechanism based on the hierarchical social networks. In the hierarchical social network, we firstly learn the attitudes of participants for the opinions of decision makers. Inspired by the thought of three-way decision, we successfully identify the valid followers, hesitant followers and invalid follower of decision makers from the participants with the aid of the trust and consensus information. According to the character of followers, we compute the real influence of decision makers. Then, we discuss the consensus model. In the adjustment process of GDM of social network, the inconsistent decision makers tend to adjust evaluation based on the evaluation of the most influential decision makers. Thus, we introduce the maximum closeness degree between the inconsistent decision maker and the most influential decision maker to construct the minimum adjustment consensus model. Besides, we extend the minimum adjustment consensus model by considering the various unit cost and limited adjustment budget. In light of the consensual loss information and Bayesian decision theory, we determine the classification rules of three-way group decision. Finally, we use an example of the medical equipment selection of purchasing department to elaborate and validate our proposed method.

Do media portrayals and social consensus information impact anti-fat attitudes and support for anti-weight discrimination laws and policies?

Although weight stigma and discrimination are associated with increased body dissatisfaction and eating disorder risk, reduced opportunities, and poorer well-being, there are few legal protections for such discrimination in the U.S. We addressed one barrier to enacting protective legislation – public attitudes toward anti-weight discrimination laws – by assessing the impact of media representations of fatness and information about peer attitudes. Using a 2 × 2 experimental design, participants (N = 525) completed baseline assessments of political conservatism and weight bias and were randomly assigned to view fat-negative or fat-positive media content that was ostensibly supported or not supported by their peers, followed by questionnaires assessing fat phobia and legislative attitudes. Two-way ANCOVAs controlling for baseline weight bias and political conservatism indicated a significant effect for media framing, with greater fat phobia and less support for anti-discrimination laws and policies among those who viewed the fat-negative frame; however, effects for ostensible peer support and interaction effects were not significant. These preliminary findings suggest that efforts to shift media rhetoric may enhance support for anti-weight discrimination laws. Future research should investigate other barriers to anti-discrimination legislation and estimate their impact on body dissatisfaction, eating disorder risk, and other indicators of population health.

Distributed Iterative Learning Formation control for Nonholonomic Multiple Wheeled Mobile Robots with Channel Noise

In this paper, we studied the robust formation control issue of multiple non-holonomic wheel mobile robots (WMRs) with nonlinear characteristics and considered the channel noise and switching communication topology, a distributed iterative learning formation control (DILFC) scheme using information interaction between robots is proposed. Firstly, the formation tracking error with consensus information is constructed, and the relationship between formation error and channel noise is obtained from the nonlinear system model of mobile robot. Next, the controller is designed based on the prediction and the current learning term between robots, and the switching topology is introduced into the formation algorithm in the form of piecewise function. The sufficient condition and norm upper bound for the formation tracking stability of the system are obtained by theoretical analysis. The results show that although the channel noise accumulates in both the time domain and iteration domain, the validity of formation tracking can be guaranteed by adjusting the sampling time of the system. To illustrate the effectiveness of the proposed scheme, numerical simulation results of a group of WMRs are presented.

Credible Peer-to-Peer Trading with Double-Layer Energy Blockchain Network in Distributed Electricity Markets

Blockchain-based peer-to-peer (P2P) energy trading is one of the most viable solutions to incentivize prosumers in distributed electricity markets. However, P2P energy trading through an open-end blockchain network is not conducive to mutual credit and the privacy protection of stakeholders. Therefore, improving the credibility of P2P energy trading is an urgent problem for distributed electricity markets. In this paper, a novel double-layer energy blockchain network is proposed that stores private trading data separately from publicly available information. This blockchain network is based on optimized cross-chain interoperability technology and fully considers the special attributes of energy trading. Firstly, an optimized ring mapping encryption algorithm is designed to resist malicious nodes. Secondly, a consensus verification subgroup is built according to contract performance, consensus participation and trading enthusiasm. This subgroup verifies the consensus information through the credit-threshold digital signature. Thirdly, an energy trading model is embedded in the blockchain network, featuring dynamic bidding and credit incentives. Finally, the Erenhot distributed electricity market in China is utilized for example analysis, which demonstrates the proposed method could improve the credibility of P2P trading and realize effective supervision.

Not out of MY bank account! Science messaging when climate change policies carry personal financial costs

We suggest that policies will be less popular when individuals personally have to pay for them rather than when others have to pay (i.e., a Not Out of My Bank Account or NOMBA effect). Dual process models of persuasion suggest that personally having to pay would motivate scrutiny of persuasive messages making it essential to use effective science communication tactics when using climate science to support climate change policies. A pilot experiment (N = 186) and main study (N = 758) support a NOMBA effect with less policy support (Pilot study) and lower recommended fees (Main study) for a policy that would require participants, rather than another group, to pay a fee for community solar panels. Consistent with dual process models and suggesting systematic processing, only when participants would have to pay the fee, messages using strong (vs. weak) science communication tactics increased support for policies (Pilot study), increased the favorability of thoughts about the policy (Main study) and these thoughts subsequently predicted policy support (Main study). Inconsistent with propositions that information about expert sources would be a heuristic or bolster science messages, expert consensus information did not influence thoughts or policy support in any study condition. Efforts to understand climate change policy support would benefit from attending to research on dual process models of persuasion, including understanding how different types and degree of outcome relevance can alter how people process science information used to bolster support for climate change policies.

Multi-view learning methods with the LINEX loss for pattern classification

Multi-view learning concentrates on leveraging the consensus and complementarity information among multiple distinct feature representations to improve the performance. Most multi-view learning models deal with two main issues. Firstly, how to fully exploit the view-agreement and view-discrepancy poses a major challenge. Secondly, how to design a general multi-view model is indispensable. By inheriting the asymmetric merit of LINEX loss, we propose a general multi-view LINEX SVM framework, which includes two models called MVLSVM-CO and MVLSVM-SIM. They can not only use LINEX loss function to flexibly distinguish the error-prone samples of both classes, but also take advantage of the consistency and the complementarity of distinct views in multi-view scenario. An iterative two-step strategy is adopted to solve the optimization problems efficiently. Furthermore, we theoretically analyze the view-consistency and generalization capability of the proposed models by using Rademacher complexity. The extensive experiments confirm the effectiveness of MVLSVM-CO and MVLSVM-SIM.

Distributed moving horizon state estimation for sensor networks with low computation capabilities

This paper focuses on distributed state estimation for sensor network observing a discrete-time linear system. The provided solution is based on a Distributed Moving Horizon Estimation (DMHE) algorithm considering a pre-estimating Luenberger observer in the formulation of the local problem solved by each sensor. This leads to reduce the computation load, while preserving the accuracy of the estimation. Moreover, observability properties of local sensors are used for tuning the weights related to consensus information fusion built on a rank-based condition, in order to improve the convergence of the estimation error. Results obtained by Monte Carlo simulations are provided to compare the performance with existing approaches, in terms of accuracy of the estimations and computation time.

Using a News Article to Convey Climate Science Consensus Information

The current study investigates how people respond to a climate science consensus statement embedded within a news article. Participants ( N = 1,048) were randomly assigned to read a news article about climate change, read the same article with a scientific consensus message included, read a simple consensus statement, or a control condition. Participants in consensus conditions had increased perceptions of scientific agreement compared with those who did not receive consensus information. Moreover, the article was similarly effective as an overt consensus statement. However, neither consensus statement affected other climate change attitudes, suggesting the effect may be limited to consensus perceptions.

A Pandemic of Misbelief: How Beliefs Promote or Undermine COVID-19 Mitigation

Sustained and coordinated social action is needed to combat the spread of the novel coronavirus disease 2019 (COVID-19). Health practitioners and governments around the world have issued recommendations and mandates designed to reduce the transmission of COVID-19 by influencing the social behaviors of the general public. Why and when are some people unwilling to take action to protect themselves and others from the effects of this public health crisis? We find that belief in COVID-19 consensus information (by the self or perceptions of scientists’ beliefs), are consequential predictors of COVID-19 mitigation behaviors. Importantly, support for COVID-19 conspiracy theories predicted decreased, whereas perceived understanding of COVID-19 predicted increased, belief in COVID-19 consensus information. We also implemented an Illusion of Explanatory depth paradigm, an approach to examining knowledge overestimation shown to reduce confidence in one’s understanding of complex phenomena. By requiring participants to elaborate upon COVID-19 conspiracies, we experimentally increased understanding of these theories, which led, in turn, to ironic increases in support for the conspiracy theories and undermined perceived understanding of COVID-19 information for a notable portion of our participants. Together, our results suggest that attention given to COVID-19 conspiracies may be misguided; describing or explaining the existence of COVID-19 conspiracies may ironically increase support for these accounts and undermine knowledge about and willingness to engage in COVID-19 mitigation.

DEA cross‐efficiency ranking method considering satisfaction and consensus degree

AbstractCross‐efficiency evaluation in data envelopment analysis is an effective way to rank decision‐making units (DMUs). However, different cross‐efficiency evaluation models derived from different perspectives generate different cross‐efficiency rankings. The information resulting from the various perspectives may be valuable and should not be ignored. In this paper, we propose an innovative composite method for ranking DMUs by calculating the Shannon entropy of the obtained cross‐efficiency scores derived from the perspectives of satisfaction and consensus. Also, we adopt grey incidence analysis to compare the rankings of different cross‐efficiency models. The calculation procedure using Shannon entropy and grey incidence analysis is illustrated on an example to generate the composite ranking result and compare it to other cross‐efficiency model rankings. The cross‐efficiency ranking using both satisfaction and consensus information provides a new comprehensive perspective in group evaluation. A practical example is used to show that the cross‐efficiency results obtained from the composite perspective of satisfaction and consensus should be widely accepted in practical decision‐making.

Express acceptance or feel anxiety? The effects of trait-related information about foreign residents on endorsement of multiculturalism

Apart from direct contact, people also familiarize themselves with other groups through indirect information. For example, as previous studies have revealed, indirect information on personality traits of outgroup members have influences on intergroup attitudes. Extending existing work, the current study specifically examined the effects of such indirect information on people’s endorsement of multiculturalism, a set of attitudes regarding the attainment of harmonious coexistence among diverse cultural groups. Based on the Stereotype Content Model, two experiments were conducted to investigate how media coverage and social consensus information about moral, social and competence-related traits of foreign residents in China would affect local people’s endorsement of multiculturalism. In experiment 1, when participants learned from the news that foreigners in China possessed high competence, they expressed more appreciation towards the cultural diversity that foreigners brought to the local society, but less support for foreign residents maintaining their own cultural traditions. In experiment 2, participants who learned that other local people perceived foreign residents as having high sociability, morality or competence showed more appreciation towards cultural diversity and more support for maintaining such diversity. Furthermore, in both experiments, perceived anxiety played a mediating role in how indirect trait-related information affected local people’s support for maintaining cultural diversity. Lastly, possible explanations for some unexpected findings and implications for future research are discussed.

Decision making with incomplete interval multiplicative preference relations based on stochastic program and interval category

• The risk attitude of DM is introduced to define a new consistency index of IMPR. • The most pessimistic and optimistic acceptably consistent IMPRs are obtained. • A stochastic programming model is built to generate the priority weights. • A 0–1 mixed integer program is established to determine the DMs’ weights. • Derive the adjusted DMs’ weights with the group consensus and category information. Interval multiplicative preference relations (IMPRs) have been widely used in decision making for their ability to efficiently express the uncertainty of information. This paper investigates the decision making with incomplete IMPRs. First, a new consistency index of IMPR is defined. By minimizing the consistency index, the missing values in an incomplete IMPR can be estimated. Subsequently, considering the risk attitude of decision-makers (DMs), two optimization models are constructed to obtain the most pessimistic and optimistic acceptably multiplicative consistent IMPRs, respectively. Based on the triangular distribution of intervals, a stochastic programming model is built to derive the priority weights from an acceptably multiplicative consistent IMPR. Thus, a new method is put forward for individual decision making with an incomplete IMPR. To reach maximum group support degree, a 0–1 mixed integer programming model is established to determine DMs’ weights for group decision making with IMPRs. Considering the category information of the individual intervals and collective intervals, the adjusted DMs’ weights are defined. The individual IMPRs are integrated into the collective IMPR. The ranking of alternatives is generated by the collective IMPR. Two real-life examples are demonstrated to validate the proposed methods. A decision support system based on the proposed methods is designed.

Group Decision-Based Construction of Scenarios for Multicriteria Analysis in Conditions of Uncertainty on the Basis of Quantitative and Qualitative Information

This study examines a general scheme of multicriteria decision-making under uncertainty based on quantitative and qualitative information. In its original form, the general scheme uses qualitative information at the final decision stage, only if the quantitative information is unable to generate unique solutions. Considering that many problems require objectives based on qualitative information at all decision stages, the results of Ramalho et al. (2019) can provide such solutions. However, the processing of qualitative information in Ramalho et al. (2019) is based on applying a very simple approach to aggregating individual preferences without analyzing any type of consensus information. Therefore, the present study sets out to show that representative combinations of initial data, states of nature or scenarios can be constructed by using qualitative information directly, based on aggregating individual preferences after achieving the necessary consensus. In particular, a new scheme for consensus construction is proposed. Using this scheme enables negative points inherent in traditional approaches to be avoided and uncertainty levels to be reduced when estimating coefficients of objective functions based on qualitative information. An illustrative example is presented to demonstrate the results that the new proposal can achieve.