Revealing the underlying causal mechanisms in the real world is crucial for scientific and technological progress. Despite notable advances in recent decades, the lack of high-quality data and the reliance of traditional causal discovery algorithms (TCDA) on the assumption of no latent confounders, as well as their tendency to overlook the precise semantics of latent variables, have long been major obstacles to the broader application of causal discovery. To address this issue, we propose a novel causal modeling framework, TLVD, which integrates the metadata-based reasoning capabilities of large language models (LLMs) with the data-driven modeling capabilities of TCDA for inferring latent variables and their semantics. Specifically, we first employ a data-driven approach to construct a causal graph that incorporates latent variables. Then, we employ multi-LLM collaboration for latent variable inference, modeling this process as a game with incomplete information and seeking its Bayesian Nash Equilibrium (BNE) to infer the possible specific latent variables. Finally, to validate the inferred latent variables across multiple real-world web-based data sources, we leverage LLMs for evidence exploration to ensure traceability. We comprehensively evaluate TLVD on three de-identified real patient datasets provided by a hospital and two benchmark datasets. Extensive experimental results confirm the effectiveness and reliability of TLVD, with average improvements of 32.67% in Acc, 62.21% in CAcc, and 26.72% in ECit across the five datasets.

Risk disclosures play a crucial role in the investment decision‐making processes for investors. However, extracting relevant variables from unstructured financial text poses a nontrivial challenge. In this paper, we propose RiskBERT, a large language model (LLM) trained on financial texts and risk knowledge graphs, specifically designed for risk factors extraction. By incorporating both finance and risk knowledge, RiskBERT significantly improves the extraction of risk factors in financial texts. We evaluate RiskBERT’s performance on a labeled risk factors dataset comprising 119,153 sentences from 2400 Chinese A‐listed companies and compare it against other LLMs and automated text analysis algorithms for risk types’ classification. Our findings demonstrate that RiskBERT outperforms alternative models, particularly when the training sample size is limited. Moreover, we uncover that RiskBERT provides risk informativeness estimates in annual reports that are at least 4.5% higher than those derived from other models. These results highlight the value of RiskBERT as a powerful tool for extracting risk factors and enhancing risk analysis in finance and accounting domains.

Revealing the underlying causal mechanisms in the real world is critical for scientific and technical progress. Despite advancements over the past decades, the lack of high-quality data and the inability of traditional causal discovery algorithms (TCDA) to fully comprehend the exact semantics of variables have long been major obstacles to the broader application of causal discovery. To address this issue, this paper proposes a novel causal modeling framework, LLM-CD, which integrates the metadata-based reasoning capabilities of large language models (LLMs) with the data-driven modeling abilities of TCDA for causal discovery. LLM-CD deeply couples the reasoning abilities of LLMs at various stages of TCDA, and enhances causal discovery through an iterative process. Due to the issues of overconfidence and hallucination in LLMs, LLM-CD quantifies and analyzes its uncertainty by incorporating evidence-based deep learning theory with the assumptions of TCDA. We utilize a large-scale de-identified real patient dataset provided by a hospital, a new dataset extracted from MIMIC-IV about the same disease (lung cancer), and two benchmark datasets to comprehensively evaluate LLM-CD. Extensive experimental results confirm the effectiveness and reliability of LLM-CD, with the highest improvement of 403.93% in the Recall and 25.77% in the Ratio metric across four datasets.

Temporal graph representation learning has drawn considerable attention in recent years. Most existing works mainly focus on modeling local structural dependencies of temporal graphs. However, underestimating the inherent global structural role information in many real-world temporal graphs inevitably leads to sub-optimal graph representations. To overcome this shortcoming, we propose a novel Role-based Temporal Graph Convolution Network (RTGCN) that fully leverages the global structural role information in temporal graphs. Specifically, RTGCN can effectively capture the static global structural roles by using hypergraph convolution neural networks. To capture the evolution of nodes' structural roles, we further design structural role-based gated recurrent units. Finally, we integrate structural role proximity in our objective function to preserve global structural similarity, further promoting temporal graph representation learning. Experimental results on multiple real-world datasets demonstrate that RTGCN consistently outperforms state-of-the-art temporal graph representation learning methods by significant margins in various temporal link prediction and node classification tasks. Specifically, RTGCN achieves AUC improvement of up to 5.1% for link prediction and F1 improvement of up to 6.2% for new link prediction. In addition, RTGCN achieves AUC improvement up to 4.6% for node classification and 2.7% for structural role classification.

Choosing the right projects in renewable energy investments is very significant. Due to this issue, necessary improvements to the performance indicators of these projects should be made. However, every improvement made also leads to an increase in costs. There is a need for a priority analysis to find the most important factors affecting the selection of the right renewable energy projects. Accordingly, this study aims to evaluate critical determinants of renewable energy project selection and provide effective investment strategies for this situation with a new fuzzy decision-making model. First, the indicators of renewable energy project selection are analyzed by perception and expression-based quantum Spherical fuzzy M-SWARA. The weights of these determinants are also calculated by DEMATEL to check the reliability of the results. Secondly, the priorities of renewable energy project selection are ranked by considering perception and expression-based quantum Spherical fuzzy ELECTRE. This calculation is also made by TOPSIS methodology to measure the reliability of the findings. The main contribution of this manuscript is that perception and expression-based evaluation can be carried out in the proposed model. In this process, the facial expressions and emotions of the decision makers are considered so that the hesitancy of these people while answering these questions can be included in the evaluation process. Another important novelty is that a new decision-making model (M-SWARA) is also proposed. This new technique provides an opportunity to consider causality relationship between the criteria to reach the most significant ones. The weighting results are the same for both M-SWARA and DEMATEL approaches. This situation gives information that the findings are coherent and valid. Market analysis has the greatest value in both perception-based (0.272) and expression-based (0.259) evaluations. Owing to this analysis, it is possible to clearly understand the supply-demand balance in the market. The ranking results indicate that technical adequacy is the most significant priority alternative for the selection of the appropriate renewable energy alternatives.

Network embedding is a fundamental part of many network analysis tasks, including node classification and link prediction. The existing random walk-based embedding methods aim to learn node embedding that preserves information on either node proximity or structural similarity. However, the information on both role and community is important to network nodes. To address the shortcomings of the existing methods, this paper proposes a novel method for network embedding called the RARE, which can be used for the analysis of different types of networks and even disconnected networks. The proposed method uses the role and community information of nodes to preserve both node proximity and structural similarity in the learned node embeddings. The walks generated through the role-aware random walk can capture the role and community information of nodes. The obtained walks are input to the Skip-gram model to learn the final embedding of nodes. In addition, the RARE is extended to the CRARE that adds the sampling of high-order community members to the customized random walk so that the node’s representation can preserve more structural information of the network. The performances of the proposed methods are evaluated on multi-class node classification, link prediction, and network visualization tasks. Experimental results on different domain datasets indicate that the proposed methods outperform the baseline methods. The proposed methods can be further accelerated using parallelization in the random walk generation process.

As a novel architecture of a fuzzy decision tree constructed on fuzzy rules, the fuzzy rule-based decision tree (FRDT) achieved better performance in terms of both classification accuracy and the size of the resulted decision tree than other classical decision trees such as C4.5, LADtree, BFtree, SimpleCart and NBTree. The concept of Z-number extends the classical fuzzy number to model both uncertain and partial reliable information. Z-numbers have significant potential in rule-based systems due to their strong representation capability. This paper designs a Z-number-valued rule-based decision tree (ZRDT) and provides the learning algorithm. Firstly, the information gain is used to replace the fuzzy confidence in FRDT to select features in each rule. Additionally, we use the negative samples to generate the second fuzzy numbers that adjust the first fuzzy numbers and improve the model's fit to the training data. The proposed ZRDT is compared with the FRDT with three different parameter values and two classical decision trees, PUBLIC and C4.5, and a decision tree ensemble method, AdaBoost.NC, in terms of classification effect and size of decision trees. Based on statistical tests, the proposed ZRDT has the highest classification performance with the smallest size for the produced decision tree.

Representation learning has an important impact on the performance of machine learning methods and has been used to solve many distribution problems for numerous graphical and sequential mining tasks. While the distributions of credit data are very complex, the representations of such data are less studied. This study proposes a new representation learning approach based on a neural network called NyströmNet, which represents the credit data to benefit credit evaluation and sub-pattern analysis. The NyströmNet is developed to utilize the advantages of the Nyström method – a kernel approximation method in credit evaluation, yet overcomes its two limitations: distance distortions in kernel functions, and parameter tuning. The two main modules contained in NyströmNet, i.e., the Distance Metric Learning module and the Nyström module, can benefit each other and yield an overall optimum. Experiments using six real-life large-scale credit data showed that the AUC of the distance-based classifiers and the linear classifiers were improved by 2–11% and 2–14% with the newly generated distributions. The proposed approach also has certain practical advantages over traditional approaches because it is free from complex parameter tuning, consumes fewer memories, and is easy to utilize automatic differential frameworks such as PyTorch. The proposed approach is highly suitable for large-scale credit evaluation.

Stock Movement Prediction (SMP) aims at predicting listed companies' stock future price trend, which is a challenging task due to the volatile nature of financial markets. Recent financial studies show that the momentum spillover effect plays a significant role in stock fluctuation. However, previous studies typically only learn the simple connection information among related companies, which inevitably fail to model complex relations of listed companies in real financial market. To address this issue, we first construct a more comprehensive Market Knowledge Graph (MKG) which contains bi-typed entities including listed companies and their associated executives, and hybrid-relations including the explicit relations and implicit relations. Afterward, we proposeDanSmp, a novel Dual Attention Networks to learn the momentum spillover signals based upon the constructed MKG for stock prediction. The empirical experiments on our constructed datasets against nine SOTA baselines demonstrate that the proposedDanSmpis capable of improving stock prediction with the constructed MKG.

Bi-typed multi-relational heterogeneous graph (BMHG) is one of the most common graphs in practice, for example, academic networks, e-commerce user behavior graph and enterprise knowledge graph. It is a critical and challenge problem on how to learn the numerical representation for each node to characterize subtle structures. However, most previous studies treat all node relations in BMHG as the same class of relation without distinguishing the different characteristics between the intra-type relations and inter-type relations of the bi-typed nodes, causing the loss of significant structure information. To address this issue, we propose a novelDualHierarchicalAttentionNetworks (DHAN) based on the bi-typed multi-relational heterogeneous graphs to learn comprehensive node representations with the intra-type and inter-type attention-based encoder under a hierarchical mechanism. Specifically, the former encoder aggregates information from the same type of nodes, while the latter aggregates node representations from its different types of neighbors. Moreover, to sufficiently model node multi-relational information in BMHG, we adopt a newly proposed hierarchical mechanism. By doing so, the proposed dual hierarchical attention operations enable our model to fully capture the complex structures of the bi-typed multi-relational heterogeneous graphs. Experimental results on various tasks against the state-of-the-arts sufficiently confirm the capability of DHAN in learning node representations on the BMHGs.

Balancing the accuracy rates of the majority and minority classes is challenging in imbalanced classification. Furthermore, data characteristics have a significant impact on the performance of imbalanced classifiers, which are generally neglected by existing evaluation methods. The objective of this study is to introduce a new criterion to comprehensively evaluate imbalanced classifiers. Specifically, we introduce an efficiency curve that is established using data envelopment analysis without explicit inputs (DEA-WEI), to determine the trade-off between the benefits of improved minority class accuracy and the cost of reduced majority class accuracy. In sequence, we analyze the impact of the imbalanced ratio and typical imbalanced data characteristics on the efficiency of the classifiers. Empirical analyses using 68 imbalanced data reveal that traditional classifiers such as C4.5 and the k-nearest neighbor are more effective on disjunct data, whereas ensemble and undersampling techniques are more effective for overlapping and noisy data. The efficiency of cost-sensitive classifiers decreases dramatically when the imbalanced ratio increases. Finally, we investigate the reasons for the different efficiencies of classifiers on imbalanced data and recommend steps to select appropriate classifiers for imbalanced data based on data characteristics.

Coupled information diffusion in complex networks has been widely studied in recent years. Nevertheless, current research mainly focuses on the interaction between each information pair. In this study, we investigate the interaction of multiple types of information on multiplex networks by considering both the competition and the cooperation among them. To study the dynamic characteristics theoretically, a microscopic Markov chain approach is used to reveal the co-evolution of multiple information. Through extensive simulations, the outbreak threshold is analyzed theoretically. The results reveal that the pairwise interaction between each information pair has an obvious impact on its final outbreak scale and the diffusion threshold. Interestingly, even information that has no direct impact on the target information can affect the diffusion of the target information through indirect effects. In addition, the inhibitory effect of the competitive information and the promotion effect of the cooperative information on the target information will reach equilibrium under specific parameter space conditions. We also conduct numerical simulations on three real multiplex social networks, including two large-scale networks. Current results are beneficial for us to further understand the coupled diffusion of multiple information on multiplex social networks.

The benchmarking of smart e-tourism data management applications falls under the problem of multicriteria decision-making (MCDM). This claim is supported by three issues: 12 smart key concepts need to be considered in the evaluation, criteria importance, and data variation among these criteria. Thus, an MCDM solution is essential to overcome problem complexity. To end this, this study presents a decision-making framework on the basis of the extension of interval type 2 trapezoidal-fuzzy weighted with zero inconsistency (IT2TR-FWZIC) integrated with the Vlsekriterijumska Optimizcija I Kaompromisno Resenje (VIKOR) method for evaluating and benchmarking the smart e-tourism data management applications. Our methodology comprises two consecutive phases. In the first phase, a decision matrix is constructed using the intersection between the 12 key concepts and smart e-tourism data management applications of each category and subcategory in smart e-tourism. In the second phase, the integration of the IT2TR-FWZIC formulation and VIKOR is presented to compute the weights for the 12 key concepts and benchmark the smart e-tourism data management applications for each category. The results are as follows: (1) A clear difference is found among the criteria weights (12 smart key concepts). Specifically, the real-time criterion achieves the highest importance weight (0.098), whereas augmented reality obtains the lowest weight (0.068). The context-awareness and recommender systems have the same weight value (0.087), and the other eight criteria are distributed in between. (2) The smart e-tourism data management applications are evaluated and benchmarked effectively per category and subcategories. (3) Benchmarked applications in each category are subjected to a systematic ranking in the evaluation process. The sensitivity analysis has shown high correlation outcomes to the systematic ranking results over the 31 scenarios of criteria weight changing. Moreover, a comparative analysis of the proposed work with other existing studies is also discussed.

Urban resettlement projects involve a large number of stakeholders and impose tremendous cost. Developing resettlement plans and reaching an agreement amongst stakeholders about resettlement plans at a reasonable cost are some of the key issues in urban resettlement. From this perspective, urban resettlement is a typical large-scale group decision-making (GDM) problem, which is challenging because of the scale of participants and the requirement of high consensus levels. Observing that residents who are affected by a resettlement project often have tight social connections, this study proposes a framework to improve the consensus reaching and uses the minimum consensus cost to reduce the total cost for urban resettlement projects with more than 1000 participants. Firstly, we construct a network topology that consists of two layers to deal with incomplete social relationships amongst large-scale participants. An inner layer consists of participants whose preference similarities and trust relations are known. Meanwhile, an outside layer includes participants whose trust relations cannot be determined. Secondly, we develop a classification method to classify participants into small subgroups based on their preference similarities. We can then connect the participants whose trust relations are unknown (the outside layer) with the ones in the inner layer using the classification results. To facilitate effective consensus reaching in large-scale social network GDM, we develop a three-step approach to reconcile conflicting preferences and accelerate the consensus process at the minimum cost. A real-life urban resettlement example is used to validate the proposed approach. Results show that the proposed approach can reduce the total consensus cost compared with the other two practices used in the actual urban resettlement operations.

Distance metric learning (DML) aims to learn distance metrics that reflect the interactions between features and labels. Due to the high computational complexity, existing DML models are unsuitable for large-scale datasets. This study proposes a DML approach for large-scale problems by reducing the number of variables, utilizing sparse structures of the optimization problems, and taking advantage of large-scale computation platforms. The proposed approach treats DML as a linear space transformation problem and suggests that a full DML matrix can be approximated by a diagonal matrix in many cases. We solve the diagonal DML problem along with its ℓ1 and ℓ2 regularizations via linear and quadratic programming. To facilitate large-scale learning problems, we design a MapReduce framework to build triplets, which are encapsulations of triple data points used for the optimization problem, and develop a weighting mechanism for triplets according to their contributions to the whole distance distortion. Experiments show that the proposed approach is fast in large-scale DML applications with comparable accuracy to much more time-consuming full-matrix models. Since the approach is implemented with the Scala language based on the Spark platform, it can be used directly by productive Java applications, which makes it highly practical for large-scale datasets.

Recent multi-criteria group decision making methods focus their analysis on the experts preferences. They do not take into account the reasons why each expert has provided a specific set of preferences. In this paper, a method that introduces novel measures capable of explaining the reasons behind experts decisions is presented. A novel concept, the arguments are presented. They represent the experts have for maintaining a certain position in the debate. Several measures related to the arguments are proposed. These new argumentation measures, along with consensus measures, help us to get a clear idea about how and why a specific resolution has been reached. They help us to determine which is the most influential expert, that is, the expert whose contributions to the debate have inspired the rest. Also, the proposed method allows us to determine which are the arguments that most of the experts have followed. A clear overview about how the debate is evolving in terms of arguments is also provided. The novel presented analysis indicate how the experts change their opinions in every round and what was the reason for it, which changes have occurred between rounds and they also provide global analysis results.

The problem of deriving the priority vector from a pairwise comparison matrix is at the heart of multiple-criteria decision-making problems. Existing prioritization methods mostly model the inconsistency in relative preference–the ratio of two preference weights–by allowing for a small deviation, either additively or multiplicatively. In this study, we alternatively allow for a deviation in each of the two preference weights, which we refer to as the relative deviation interconnection. Under this framework, we consider both relative additive and multiplicative deviation cases and define two types of norms capturing the magnitudes of the deviations, which gives rise to four conic programming models for minimizing the norms of the deviations. Through the model structures, we analyze the signs of the deviations. This further allows us to establish the expressiveness of our framework, which covers the logarithmic least-squares method and the goal-programming method. Using numerical examples, we show that our models perform comparably against existing prioritization methods, efficiently identify unusual and false observations, and provide further suggestions for reducing inconsistency.

In malicious URLs detection, traditional classifiers are challenged because the data volume is huge, patterns are changing over time, and the correlations among features are complicated. Feature engineering plays an important role in addressing these problems. To better represent the underlying problem and improve the performances of classifiers in identifying malicious URLs, this paper proposed a combination of linear and non-linear space transformation methods. For linear transformation, a two-stage distance metric learning approach was developed: first, singular value decomposition was performed to get an orthogonal space, and then a linear programming was used to solve an optimal distance metric. For nonlinear transformation, we introduced Nyström method for kernel approximation and used the revised distance metric for its radial basis function such that the merits of both linear and non-linear transformations can be utilized. 33,1622 URLs with 62 features were collected to validate the proposed feature engineering methods. The results showed that the proposed methods significantly improved the efficiency and performance of certain classifiers, such as k-Nearest Neighbor, Support Vector Machine, and neural networks. The malicious URLs’ identification rate of k-Nearest Neighbor was increased from 68% to 86%, the rate of linear Support Vector Machine was increased from 58% to 81%, and the rate of Multi-Layer Perceptron was increased from 63% to 82%. We also developed a website to demonstrate a malicious URLs detection system which uses the methods proposed in this paper. The system can be accessed at: http://url.jspfans.com.

Nowadays, in the social network–based decision-making processes, like the ones involved in e-commerce and e-democracy, multiple users with different backgrounds may take part and diverse alternatives might be involved. This diversity enriches the process, but at the same time, increases the uncertainty of opinions. This uncertainty can be considered from two different perspectives: (i) the uncertainty in the meaning of the words given as preferences, that is, motivated by the heterogeneity of the decision makers; and (ii) the uncertainty inherent to any decision-making process that may lead to an expert not being able to provide all their judgments. The main objective of this study is to address these two types of uncertainty. To do so, the following approaches are proposed: First, to capture, process, and keep the uncertainty in the meaning of the linguistic assumption, the Interval Type-2 Fuzzy Sets are introduced as a way to model the experts' linguistic judgments. Second, a measure of the coherence of the information provided by each decision maker is proposed. Finally, a consistency-based completion approach is introduced to deal with the uncertainty presented in the expert judgments. The proposed approach is tested in an e-democracy decision-making scenario.

On-line platforms foster the communication capabilities of the Internet to develop large-scale influence networks in which the quality of the interactions can be evaluated based on trust and reputation. So far, this technology is well known for building trust and harnessing cooperation in on-line marketplaces, such as Amazon (www.amazon.com) and eBay (www.ebay.es). However, these mechanisms are poised to have a broader impact on a wide range of scenarios, from large scale decision making procedures, such as the ones implied in e-democracy, to trust based recommendations on e-health context or influence and performance assessment in e-marketing and e-learning systems. This contribution surveys the progress in understanding the new possibilities and challenges that trust and reputation systems pose. To do so, it discusses trust, reputation and influence which are important measures in networked based communication mechanisms to support the worthiness of information, products, services opinions and recommendations. The existent mechanisms to estimate and propagate trust and reputation, in distributed networked scenarios, and how these measures can be integrated in decision making to reach consensus among the agents are analysed. Furthermore, it also provides an overview of the relevant work in opinion dynamics and influence assessment, as part of social networks. Finally, it identifies challenges and research opportunities on how the so called trust based network can be leveraged as an influence measure to foster decision making processes and recommendation mechanisms in complex social networks scenarios with uncertain knowledge, like the mentioned in e-health and e-marketing frameworks.

In this paper, from the perspective of opinion dynamics theory, we investigate the interaction mechanism of a group of autonomous agents in an e-commerce community (or social network), and the influence power of opinion leaders during the formation of group opinion. According to the opinion's update manner and influence, this paper divides social agents within a social network into two subgroups: opinion leaders and opinion followers. Then, we establish a new bounded confidence-based dynamic model for opinion leaders and followers to simulate the opinion evolution of the group of agents. Through numerical simulations, we further investigate the evolution mechanism of group opinion, and the relationship between the influence power of opinion leaders and three factors: the proportion of the opinion leader subgroups, the confidence levels of opinion followers, and the degrees of trust toward opinion leaders. The simulation results show that, in order to maximize the influence power in e-commerce, enhancing opinion leaders’ credibility is crucial.