Search

Over the past few decades, short text classification has emerged as a critical downstream task in natural language processing (NLP). One crucial classification research issue is how to advance semantic understanding considering the short length, feature sparsity, and high ambiguity in short texts. Recently, prompt-tuning has been proposed to insert a template into the input and convert text classification tasks into equivalent cloze-style tasks. However, among most of the previous approaches, either the crafted template methods are time-consuming and labor-intensive, or automatic prompt generation methods cannot achieve satisfied performance. In this paper, we introduce a novel approach called Soft Knowledgeable Prompt-tuning for short text classification. Our method considers both the template generation and classification performance to construct prompts for label prediction. We employ five different strategies to expand the label words space for modifying soft prompts, and the integration of these strategies is used as the final verbalizer. Despite being automatic, experimental results show that our method achieved more desirable performance even than the crafted template methods, outperforming the state-of-the-art by more than 14 Accuracy points on four well-known benchmarks. • Both the automatic prompt generation and text classification performance are achieved. • 5 different strategies are used to construct verbalizer effectively and efficiently. • The comprehensive experiments evaluate the effectiveness of our proposed method. [ABSTRACT FROM AUTHOR]

High-precision segmentation of skin lesions is essential for early diagnosis of skin cancer and improved patient survival. However, this task becomes challenging due to the irregularity of skin lesion regions, significant differences in color shades, ambiguity of boundaries, and other complex interfering factors. In this study, a transformer-based multi-attention hybrid network, TMAHU-Net, is proposed to cope with the complexity of skin lesion regions. The model introduces an innovative hybrid module that fully combines the advantages of CNNs and transformers to capture both global and local feature information. We also employ a deeply separable convolutional attention module to dynamically assign attention weights, which enhances the channel and spatial dimensions of feature information. To improve contextual relevance, we introduce an extended gated external attention module. Finally, a multi-scale aggregation module is used to fuse feature information at different scales to compensate for the differences in processing images with different resolutions. The model performs well in dealing with irregular, color shades, fuzzy boundaries, and complex interference lesion regions, which provides strong support for doctors to accurately determine complex lesion regions. Critically evaluated on three publicly available datasets, ISIC2016, ISIC2017, and PH2, TMAHU-Net achieved significant improvements over state-of-the-art methods, with F1 scores and IoU scores increasing by 1.01% and 1.01%, 1.22% and 1.40%, and 0.92% and 1.50%, respectively. • Transformer-based multi-attention hybrid network for skin lesion segmentation. • Focus on segmentation of complex lesion areas. • The LCT module integrates CNN and Transformer strengths. • Multiple attention mechanisms enable the model to focus on the target region. [ABSTRACT FROM AUTHOR]

This paper studies the resource-constrained project scheduling problem where uncertain durations are caused by various disruptions with unknown probabilistic functions. To achieve a balance between risk control and computational efficiency, we employ distributionally robust optimization minimizing the worst-case Conditional Value-at-Risk when the upper bounds of durations are not explicitly stipulated. We build a path-based two-stage model over a moment-based ambiguity set. The first stage explicitly generates paths satisfying precedence and resource constraints without any information of uncertainties, while the second stage determines the critical paths over the ambiguity set. Thus, we employ an enhanced Benders decomposition to solve this model where a weighted Benders cut may discriminatingly remove arcs on the critical paths. To ease the computation burden, we develop priority constraints for resource flow and a path elimination method to handle the paths generated by the first stage. A numerical experiment shows that this Benders cut together with the path elimination method can solve more complicated instances efficiently. The result further indicates that this model performs better than the classical model of robust optimization when providing upper bounds for durations. We extract some managerial insights that suggest important guidelines for controlling risk in project scheduling. [ABSTRACT FROM AUTHOR]

The selection of antivirus masks is an important problem in the context of the ongoing COVID-19 pandemic. Multiple attribute decision-making (MADM) algorithmic approaches can be used to evaluate and compare different masks based on multiple criteria, such as effectiveness, comfort, and cost. An aggregation of interval-valued multi-fuzzy hypersoft sets provides a flexible framework for handling uncertainty and imprecision in the MADM process. This approach allows for the integration of multiple sources of information such as expert opinions and empirical data, and considers the different levels of uncertainty and ambiguity associated with each criterion. By using the matrix-manipulated aggregation of interval-valued multi-fuzzy hypersoft sets like the induced fuzzy matrix, α -level matrix, threshold matrix, and mid-threshold matrix, an algorithm is proposed for the optimal selection of material for manufacturing antivirus masks. The robustness of the algorithm is maintained by following simple computation-based stages that enable a wide range of multidisciplinary readers to understand the idea vividly. By using this algorithm, it is possible to improve the accuracy and reliability of the decision-making process and to better balance the trade-offs between the different criteria, i.e., the computed results of the proposed algorithm and the structural aspects of the proposed approach are both compared with some relevant existing structures. Computation-based and structural comparisons are presented to assess the adaptability and reliability of the study. The first one is meant to check reliability, while the second is meant to check flexibility. In both cases, however, the presented approach yields the required standard. By comparing the prospective structure to the relevant developed model, the implications of the proposed framework are explored. [ABSTRACT FROM AUTHOR]

With the constant changes in the decision-making environment, various limitations have emerged in traditional decision-making methods, such as the inability to simultaneously consider multiple factors such as randomness, hesitation, and ambiguity, thus limiting a comprehensive grasp of the decision-maker's cognitive inputs. To address the above problems, we propose a multi-criteria decision-making (MCDM) model based on a new fuzzy set (FS). Firstly, we define a new Z hesitant fuzzy linguistic term set (ZHFLTS) and propose a visualization metric for mapping ZHFLTS into T-spherical fuzzy (T-SF) space to simplify the computation. The relevant properties of ZHFLTS are further investigated. In order to accurately measure the uncertainty of FSs, we propose the probabilistic-hesitation-fuzzy (PHF) entropy and use it as a basis to derive the potential probability distribution using the fruit fly optimization algorithm (FOA). Subsequently, an MCDM model is established based on the proposed ZHFLTS, visualization metric, and PHF entropy. The model can better solve complex decision-making problems such as TCM through example studies and comparative analyses. The research in this paper aims to cope with today's increasingly complex decision-making environments, and the introduced ZHFLTS framework and the corresponding methodology provide new ways to solve complex decision problems. • ZHFLTS is defined and a mapping method to visualize ZHFLTS in T-SF space is studied. • PHF entropy is proposed based on FOA for obtaining potential probability distribution. • A Novel MCDM model based on ZHFLTS is presented. • It proves the effectiveness of the model through instances and comparative analysis. [ABSTRACT FROM AUTHOR]

Partial multi-view clustering (PMVC) is a hot topic in data mining, where each view suffers from the absence of some data. Although many PMVC algorithms have been presented with appropriate performance, to the best of our knowledge, almost all of them need to be fed with the number of clusters as hyperparameters. Moreover, these existing algorithms only create hard partitions for objects with multiple views, which are often in highly overlapping areas. This ignores ambiguity and uncertainty in object assignment, which likely leads to performance degradation. To address these issues, we propose a novel self-filling evidential clustering algorithm (SFPMEC) for the PMVC problem. Based on the shared coefficient matrix learned from a self-filling process, SFPMEC provides a human-readable chart through which the number of clusters can be easily detected. SFPMEC then derives a credal partition under the framework of evidence theory, improving the fault tolerance of clustering and gaining deeper insight into the data structure. Extensive experiments are conducted to demonstrate the benefits of adopting the self-filling process and evidence theory. Besides, our algorithm shows better performance than other state-of-the-art methods on benchmark datasets 1 1 The source code can be found in https://drive.google.com/file/d/1KMi0N8cX85YhwiNjetBwA6QeolQP5TnC/view?usp=sharing.. • The issue of detecting the cluster number is solved in a PMVC problem. • The credal partition is derived for a partial multi-view dataset. • Two optimization algorithms are proposed to solve the resulting problems. • Extensive ablation and comparison experiments are conducted. [ABSTRACT FROM AUTHOR]

• Identifying blockchain solutions to gain sustainability in the supply chains. • Offering a new application for using blockchain in the agricultural products. • Presenting a novel integrated SWARA-CoCoSo approach based on Z-numbers. • Evaluating sensitivity analysis and validation of the proposed approach results. • Providing flexible, robust, and reliable results using the developed framework. Nowadays, because of the increase in global competition and the need to pay attention to sustainable issues, achieve a competitive advantage, reduce costs, and improve Supply Chain (SC) network management, the use of blockchain and its integration have become necessary in the network. The use of Blockchain Technology (BT) leads to transparency, controllability, security, flexibility, visibility, reduction of costs, efficient processes, and achieving sustainable development goals in the SC network. To successfully implement BT in SC, it is necessary to identify and evaluate the operational solutions to solve the challenges and obstacles of BT. Therefore, this study aims to identify and evaluate BT implementation solutions to achieve sustainability in the SC. The novelties and contributions of this research include identifying blockchain solutions to achieve sustainable development goals in the SC, proposing a new application for the agricultural products SC to utilize blockchain, and presenting a novel hybrid decision procedure including the Step-wise Weight Assessment Ratio Analysis (SWARA) and the Combined Compromise Solution (CoCoSo) using on Z-numbers for evaluating solutions. The proposed approach can deal with ambiguity in the complex environment of the decision due to the consideration of reliability and uncertainty in the data, reducing the number of pairwise comparisons, low inconsistency rate, reducing the complexity of calculations, ease of implementation, no limit in the number of criteria, no dependence on predetermined scales, and providing compromise solutions based on different opinions of experts. To examine the results, sensitivity analysis, and validation were performed. The findings showed that for the successful implementation of blockchain in the agricultural SC, the infrastructure of this technology must be created and integrated with all levels of the SC, and using BT must be provided through the use of sustainable methods. By developing areas of cooperation and participation through the creation of facilitating rules between different partners in the SC, collective investments can be made in the development of the infrastructure. [ABSTRACT FROM AUTHOR]

The continuous development of advanced manufacturing has triggered a strong demand for technological service platforms. To better support the development of the technological service platform, it is necessary to conduct a performance evaluation for the technological service platform. However, the existing evaluation methods fail to fully consider the ambiguity and uncertainty of decision-makers' judgments. Meanwhile, conventional methods require quite a few pairwise comparisons when determining the criteria weights. This paper proposes a novel approach that integrates rough Z-number, the best worst method (BWM), and TODIM (an acronym in Portuguese of Interactive and Multi-Criteria Decision Making) to address these issues. The proposed method can significantly reduce the number of pairwise comparisons. It also shows a significant advantage in dealing with vague information and handling the reliability of the evaluation. Moreover, the TODIM method helps capture the decision-makers' risk attitude towards uncertainty. Finally, a case study of seven technological service platforms is conducted to validate the feasibility and effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

• A novel three-stage image outpainting model is proposed. • The model predicts edge, color, and content filling in that order. • Recovery of semantic content and structure information performed well. Human beings have a natural ability to perceive invisible things around them based on materials. This paper examines the problems encountered during image outpainting, which is widely utilized in computational photography, image editing and computer graphics. Although considerable progress has been made in image outpainting, problems such as semantic ambiguity, structural confusion and poor quality still arise. Inspired by the idea that humans draw a boundary contour, paint it with color, and then fill the content, we propose a three-stage GAN model, defined as the ECPIO network. The first-stage model is an edge prediction network (EP-net) that is used to predict the edge map of missing areas. The second-stage model is a color prediction network (CP-Net) that is utilized to predict the color map in the missing area, and the third-stage model is an image outpainting network (IO-Net) that is employed to generate the outpainting results. Since edge and semantics constrain the extended area, we introduce edge and semantic information to the discriminator to guide the image generation. Our method achieved a good performance with a PSNR of 27.15 and an SSIM of 0.78 for the Cityscapes dataset. The experimental results for other public datasets also show that our method can recover accurate semantic content and accurate structure information of missing regions. In addition, our method achieves good effects in semantic coherence, structure vision, color vision and texture vision. Our method has a wide range of intelligent image processing systems, such as computer photography, image editing, and image restoration. [ABSTRACT FROM AUTHOR]

• Direct relation detection using the hierarchical structure of relations. • Relation detection without using candidate relations. • Using POS tag of words of questions to compensate for the lack of data. • One-hot encoding at the character level is used to manage out-of-vocabulary words. This study addresses the problem of relation detection for answering single-relation factoid questions over knowledge bases (KBs). In this kind of questions, the answer is obtained from a single KB fact in the form of subject-predicate-object. Conventional fact extraction methods have two steps: entity linking and relation detection, in which the output of the entity linking is used by the relation detection step to first find candidate relations, and then choose the best relation from candidate relations. Such methods have difficulties with the relation detection if there is an error or ambiguity in the entity linking step. This paper explores the relation detection task without the entity-linking step utilizing the hierarchical structure of relations and an out-of-box POS tagger. As relations are of different levels of abstraction, the proposed solution uses multiple classifiers in pipeline, each of which uses separate BiGRU neural networks fed with questions embedded with one-hot encoding at the character level. Besides, to increase the accuracy of the proposed model and to avoid the need for large amounts of training data, after each word of the question, its POS tag is inserted before feeding the network. The experimental results show that the accuracy of the proposed solution for the direct relation detection is 89.5%. In addition, the proposed solution can be used for the indirect relation detection whose accuracy is 96.3%, which is higher than state-of-the-art relation detection techniques. Finally, the positive effects of using POS tags have been examined. [ABSTRACT FROM AUTHOR]

Semantic image segmentation is challenging due to the large intra-class variations and the complex spatial layouts inside natural scenes. This paper investigates this problem by designing a new deep architecture, called multiscale sum-product network (MSPN), which utilizes multiscale unary potentials as the inputs and models the spatial layouts of image content in a hierarchical manner. That is, the proposed MSPN models the joint distribution of multiscale unary potentials and object classes instead of single unary potentials in popular settings. Besides, MSPN characterizes scene spatial layouts in a fine-to-coarse manner to enforce the consistency in labeling. Multiscale unary potentials at different scales can thus help overcome semantic ambiguities caused by only evaluating single local regions, while long-range spatial correlations can further refine image labeling. In addition, higher orders are able to pose the constraints among labels. By this way, multi-scale unary potentials, long-range spatial correlations, higher-order priors are well modeled under the uniform framework in MSPN. We conduct experiments on two challenging benchmarks consisting of the MSRC-21 dataset and the SIFT FLOW dataset. The results demonstrate the superior performance of our method comparing with the previous graphical models for understanding scene images. [ABSTRACT FROM AUTHOR]

The continuous development of e-commerce recommendation systems enables people to better obtain the products they need, saving economic and time costs. When recommending products, massive online user reviews can objectively reflect the efficacy of products and provide more practical reference value. However, due to the fuzziness of expression, there exists great uncertainty about the information in user reviews. The probabilistic linguistic term set (PLTS) is a useful tool that can help solve the information ambiguity and distinguish the importance of different terms. This paper regards PLTS as a data statistical tool to depict the information of user reviews. Based on the theory of PLTS, this paper proposes a hotel recommendation algorithm. First, we analyse the statements of online user hotel reviews with the aid of Jieba and TF-IDF. Second, the statements of reviews are translated into PLTSs and stored in the evaluation matrix. Third, we adopt the maximum deviation method to derive the attribute weights of hotels. Finally, we use probabilistic linguistic cosine similarity to calculate the similarity degrees of hotels and rank the recommended hotels according to similarity. With respect to different hospital products of users, the recommendation orders of hotels are given according to their degrees of similarity. To verify the effectiveness and superiority of the proposed algorithm, this paper selects 10 hotels in the city of Zhengzhou for case application and conducts a comparative analysis with other recommendation methods. [ABSTRACT FROM AUTHOR]

This paper focuses on an emergency rescue network design problem in response to disasters under uncertainty. Considering the limited distribution information of the uncertain demands extracted from the historical data, we use the mean absolute deviation (MAD) that can derive tractable reformulations and better capture outliers and small deviations, to construct a MAD-based ambiguity set. A distributionally robust optimization model is proposed with the objective of minimizing the preparedness cost and the expected penalty cost of demand shortage under the worst-case distribution over the ambiguity set. We analyze the constructed model and provide some features such as the theoretical bounds of the objective value. For large-scale cases, we reformulate the knotty model using the linear decision rule to obtain tight and tractable problems. Computational experiments verify that the out-of-sample performance of the proposed model is better than that of the stochastic optimization model, especially for extreme cases. The MAD-based ambiguity set combined with the approximation technique can reduce the solution time and obtain high-quality solutions. Moreover, the results show that the amount of data has a significant effect on model performance. These results provide references for decision-makers in the practice of emergency response network design. • Emergency response network distributionally robust optimization. • Mean absolute deviation-based ambiguity set. • Reformulation based on the linear decision rule. [ABSTRACT FROM AUTHOR]

Traditional Importance–Performance Analysis assumes the distribution of a given set of attributes in four sets, “Keep up the good work”, “Concentrate here”, “Low priority” and “Possible overkill” , corresponding to the four possibilities, high–high, low–high, low–low and high–low, of the pair performance–importance. This can lead to ambiguities, contradictions or non-intuitive results, especially because the most real-world classes are fuzzy rather than crisp. The fuzzy clustering is an important tool to identify the structure in data, therefore we apply the Fuzzy C -Means Algorithm to obtain a fuzzy partition of a set of attributes. A membership degree of every attribute to each of the sets mentioned above is determined, against to the forcing categorization in traditional Importance–Performance Analysis. The main benefit is related with the deriving of the managerial decisions which become more refined due to the fuzzy approach. In addition, the development priorities and the directions in which the effort of an economic or non-economic entity would be useless or even dangerous are identified on a rigorous basis and taking into account only the internal structure of the input data. [ABSTRACT FROM AUTHOR]

The fuzzy DEMATEL (Decision Making Trial and Evaluation Laboratory) method is one of the accepted decision-making methods in uncertain environment. The Pythagorean fuzzy set is a generalized concept of fuzzy set and intuitionistic fuzzy set. In this paper, we develop the Pythagorean fuzzy set based DEMATEL method and apply it to solve the supplier selection problem in sustainable supply chain management. We consider the Pythagorean fuzzy set for handling uncertain and incomplete information while selecting the criteria of a supplier and dealing with ambiguity of human decisions. Considering the independence of the criteria, we put forward the proposed method which gives mutual relationships among the criteria, and identifies cause–effect components of the system. The proposed method is illustrated numerically based on data collected from a group of professional personnel. • Extension of DEMATEL method with pythagorean fuzzy sets. • Pythagorean fuzzy sets for dealing with uncertain and incomplete information. • Solution procedure using Trapezoidal Pythagorean fuzzy number. • Solution of supplier selection problem in sustainable supply chain management. [ABSTRACT FROM AUTHOR]

Abstract: Association rule mining and classification are important tasks in data mining. Using association rules has proved to be a good approach for classification. In this paper, we propose an accurate classifier based on class association rules (CARs), called CAR-IC, which introduces a new pruning strategy for mining CARs, which allows building specific rules with high confidence. Moreover, we propose and prove three propositions that support the use of a confidence threshold for computing rules that avoids ambiguity at the classification stage. This paper also presents a new way for ordering the set of CARs based on rule size and confidence. Finally, we define a new coverage strategy, which reduces the number of non-covered unseen-transactions during the classification stage. Results over several datasets show that CAR-IC beats the best classifiers based on CARs reported in the literature. [Copyright &y& Elsevier]

This paper describes our Japanese–Chinese information retrieval system. Our system takes the “query-translation” approach. Our system employs both a more conventional bilingual Japanese–Chinese dictionary and Wikipedia for translating query terms. We propose that Wikipedia can be used as a good NE bilingual dictionary. By exploiting the nature of Japanese writing system, we propose that query terms be processed differently based on the forms they are written in. We use an iterative method for weight-tuning and term disambiguation, which is based on the PageRank algorithm. When evaluating on the NTCIR-5 test set, our system achieves as high as 0.2217 and 0.2276 in relax MAP (mean average precision) measurement of T-runs and D-runs. [Copyright &y& Elsevier]

Abstract: Process monitoring phase is one of the service-oriented business process (SOBP) lifecycle phases. Traditional process monitoring approaches have been only achieved at the syntactic level of the process monitoring contexts, which causes the communication problems such as ambiguous understandings and divergent interpretations. To solve the problems, the process monitoring should be achieved at the semantic level as well as at the syntax level of the process monitoring context. In order to support semantic monitoring operations, an ontology-based monitoring framework for the SOBP execution is suggested in this paper. The suggested framework combines a BPEL4WS process model with the semantic monitoring context which is expressed with OWL. [Copyright &y& Elsevier]

• An algorithm to mining multi-community structures of crowdsourced workers. • A method for the worker characteristics model. • Considering workers' ability, activity, preference and community attributes. • Two optimization strategies to alleviate the recommended cold start problem. Crowdsourcing as an important computing resource for the Crowd-based Cooperative, crowdsourcing workers, due to the complexity of their personnel composition and the ambiguity of group characteristics, make it difficult to accurately recommend suitable worker groups for the task. Recommending crowdsourcing tasks to suitable workers can greatly improve the efficiency and quality of the execution of crowdsourcing tasks. The current crowdsourcing recommendation algorithm is based on the single characteristics of the workers, ignoring the multi-dimensional characteristics of the workers and the community characteristics of the crowd. This paper proposes a worker group recommendation method based on multi-community collaboration (GRMBC) by utilizing the worker characteristics information extracted from crowdsourcing platform. Based on the characteristics of workers'reputation, preference and activity, the method divides the worker group into several characteristics communities with similar behavior to discover the potential multi-community structure in the crowdsourcing worker group, and then selects Top-N worker group for recommendation through the interaction among multi-communities. At the same time, we also proposed two mitigation strategies for the cold start problem of data in crowdsourcing. This paper uses the public data collected by AMT to do the experiments, and compares the aggregation results of the recommendations generated by different algorithms. The results show the recommendations generated by the GRBMC algorithm proposed in this paper performs the best comprehensively. When the GRBMC algorithm that considers both worker attributes and communi-ty characteristics has an accuracy improvement of 0.03 ~ 0.04 compared with not using the recommendation algorithm on each index, the accuracy of 0.01 ~ 0.02 is improved com-pared with the method considering single characteristic. Moreover, compared with the individual recommendation method of the workers, the recommendation method of the workers' group is more in line with the platform requirements, and can well reflect the wisdom of crowd wisdom. [ABSTRACT FROM AUTHOR]

Business rules give rise to an important set of requirements on any system being developed or procured for an enterprise. While most of the work done in this area focuses on identifying and documenting business rules, we have proposed a methodology that addresses several aspects of the business rules lifecycle: acquisition, deployment and evolution. The methodology assumes that business rules are expressed in terms of business concepts and corporate knowledge that are captured in a high level architecture. The architecture proposed consists of three interconnected components: the enterprise model, the business rules model and the decision support model. This approach permits a greater variety of rules to be specified while providing an opportunity to automate the production of deployable business rules. The ability to deal with the inconsistent and ambiguous rules is crucial in capturing the conflicting requirements placed on the operation of any large scale enterprise. This paper presents a flexible deployment of business rules, which not only supports decision making in the face of conflicting requirements, but also the evolution of those requirements in the face of changing regulatory environments, competitive markets and corporate goals. [Copyright &y& Elsevier]

Previous studies have demonstrated that images are of great importance in attracting people's attention and motivating them to take action. Various attributes (e.g., colors, aesthetics, and embedded objects) related to images are considered driving factors. Among which emotions in images, in particular, play a critical role in stimulating individuals, based on the Stimulus–Organism–Response theory. Consequently, many researchers put great efforts to understand image emotions, ranging from developing theoretical models to a broad spectrum of applications. Due to the complex and unstructured characteristics of images, identifying image emotions is challenging. Although some significant progress in image emotion classification has been achieved, inherent constraints still remain unaddressed. For example, acquiring a sufficiently large amount of labeled data to train a good model is costly and inevitably requires lots of human efforts. Besides, building a generalized model applicable to different datasets still needs a deep exploration. Image emotions are very subjective, which also makes such a classification task difficult. This paper proposes a general meta-learning framework for the few-shot image emotion classification, called Meta-IEC. Meta-IEC provides the capability of: (i) adapting to a similar dataset but new classes that have not been encountered before, and (ii) generalizing to a completely different dataset where emotion classes are unseen in the training dataset and only very few labeled images are available. Meta-IEC is also able to capture the uncertainty and ambiguity during the meta-testing, where we implement a hierarchical Bayesian graphical model to understand latent relationships among various parameters between meta-training and meta-testing. Extensive experiments on three commonly used datasets empirically demonstrate the superiority of our method over several state-of-the-art baselines. For example, our meta-learning based model can achieve performance improvement up to 5+%. We also provide some managerial implications on parameter sensitivity and label selection of meta-training and meta-testing. • Formalize image emotion classification as a few-shot learning problem. • Explore both classical and generalized few-shot image emotion classification tasks. • Propose a hierarchical Bayesian model to capture uncertainty and ambiguity. • Transfer knowledge from label-rich datasets to those with weak supervision. [ABSTRACT FROM AUTHOR]

• Develop a novel Fuzzy Network DEA approach using adjustable possibilistic programming. • Reviews real-world applications of FNDEA literature. • The proposed fuzzy network DEA model is flexible, adjustable, general, and applicable. • The FNDEA method is capable to rank of Two-stage DMUs under fuzzy data. • The presented approach is applied to the performance evaluation of investment firms. This paper presents a novel approach for performance appraisal and ranking of decision-making units (DMUs) with two-stage network structure in the presence of imprecise and vague data. In order to achieve this goal, two-stage data envelopment analysis (DEA) model, adjustable possibilistic programming (APP), and chance-constrained programming (CCP) are applied to propose the new fuzzy network data envelopment analysis (FNDEA) approach. The main advantages of the proposed FNDEA approach can be summarized as follows: linearity of the proposed FNDEA models, unique efficiency decomposing under ambiguity, capability to extending for other network structures. Moreover, FNDEA approach can be applied for ranking of two-stage DMUs under fuzzy environment in three stages: 1) solving the proposed FNDEA model for all optimistic-pessimistic viewpoints and confidence levels, 2) then plotting the results and drawing the surface of all efficiency scores, 3) and finally calculate the volume of the three-dimensional shape in below the efficiency surface. This volume can be as ranking criterion. Remarkably, the presented fuzzy network DEA approach is implemented for performance appraisal and ranking of investment firms (IFs) with two-stage processes including operational and portfolio management process. Illustrative results of the real-life case study show that the proposed approach is effective and practically very useful. [ABSTRACT FROM AUTHOR]