Recently, the metric-based prototypical methods achieves promising performance in few-shot learning (FSL) for hyperspectral image (HSI) classification. However, the existing models are easily affected by the noisy pixels of different categories around the center pixel of the patch, and tend to focus on the most representative features while ignoring other important ones, which cause the overfitting problem. Moreover, the commonly used dimension reduction operation of the feature of source and target domains inevitably results in the loss of valuable spectral information. To address these issues, we propose the mix-mask augmentation and self-reconstruction for cross-domain HSI classification. The pixel mask augmentation is introduced to enhance the sample diversity of query set and suppress the impact of noisy pixels, thus encouraging the model to discover discriminative features on a wider range. The CutMix augmentation is also adopted to generate the mixed support set and mixed prototypes, mitigating the negative impact of confusing prototypes. Furthermore, we develop the self-reconstruction module which can preserve more useful feature information during the dimension reduction for feature representation of the source and target domains. Extensive experiments on three public HSI datasets demonstrate that the proposed method achieves superior performance with fewer computational costs in comparison with the SOTA methods.

International crude oil prices are one of the important indicators in the global economy. Forecasting on crude oil prices can provide a predictive perspective for financial investment and development decision. This study explores the application of functional data analysis (FDA) techniques in the realm of crude oil price prediction, incorporating derivative information, and mixed-frequency data. The inclusion of derivative information from price trajectories is a key aspect of this study. It enriches the modeling process, offering valuable insights into rate-of-change and volatility patterns, ultimately improving predictive accuracy. In addition, the incorporation of mixed-frequency data, spanning diverse economic indicators and their respective time series, enhances the predictive accuracy of the forecasting model. To achieve a robust and interpretable decomposition of the crude oil price signal, a multivariate empirical mode decomposition (MEMD) approach is introduced. Subsequently, employing the adaptive neural fuzzy inference system to forecast submodes and aggregate them yields the ultimate prediction outcome. Empirical validation is conducted using historical Brent crude oil price datasets and robustness testing is performed using west texas intermediate (WTI) oil price data. Comparative analyses with conventional time series prediction models reveal the superiority of the proposed approach in capturing intricate temporal dynamics, irregular patterns, and abrupt changes.

Recently, convolutional neural network (CNN) and graph convolutional network (GCN) have been used widely for hyperspectral image (HSI) classification which, respectively, specialize in characterizing the local receptive feature and structure feature. However, the existing CNN-based methods cannot learn the higher-order interactions of different spectral bands. The GCN-based methods mostly used the fixed or simple graph model for feature learning. To solve the problems, we propose the dynamic hypergraph convolution and recursive gated convolution fusion network (DHCRGCFN) for HSI classification. To learn the hidden and important relations represented in the HSI data, the dynamic hypergraph convolution network (DHCN) is designed which dynamically updates the hypergraph model and captures the global spatial information of HSI. To efficiently model the high-order interactions among the high spectral dimension, the recursive gated convolution network (RGCN) is developed for progressively capturing the interactions of spectral feature. The features extracted by the two branches are adaptively fused to achieve the complementary advantages. Extensive experiments are conducted on two public HSI datasets to demonstrate the effectiveness of the proposed DHCRGCFN.

Hierarchical deep features can provide multilevel abstractions of target objects, which play an important role in target localization and classification. Determining how to effectively aggregate abstract information from different levels in RGB and thermal modalities is the key to exploiting their complementary advantages for robust RGBT tracking. However, existing RGBT tracking algorithms either focus on the semantic information of the last layer or aggregate hierarchical deep features from each modal using simple operations (e.g., summation and concatenation), which limit the capability of the multimodal tracker. To address these issues, in this paper, we propose a novel multimodal cross-layer bilinear pooling network for RGBT tracking. In our network, firstly, to boost the performance of the tracker, we use a channel attention mechanism to implement the adaptive calibration of feature channels for all convolutional layer features before realizing hierarchical feature fusion. Then, a bilinear pooling operation is performed on any two layers through the cross product, which is a second-order computation that effectively aggregates the deep semantic and shallow texture information of the target. Finally, a quality-aware fusion module is designed to aggregate the bilinear pooling features of different layer interactions between different modalities in an adaptive manner. The results of a large number of experiments on two public benchmark datasets demonstrate the effectiveness of our tracker compared with other state-of-the-art tracking methods.

In this study, the concept of linguistic Z-number fuzzy soft set ( L Z n F S S) is proposed to describe multiple uncertainties in practical decision making problems. L Z n F S S combines the concepts of fuzzy soft set, linguistic Z-number, and soft set, which could reflect both of the uncertainty in structure and the uncertainty in detailed evaluations. As an initial idea, the set operations on L Z n F S S s are put forward, the properties of such operations are also discussed. With traditional soft set based decision procedure and fuzzy soft set based decision procedure, a novel linguistic Z-number fuzzy soft set based group decision procedure is developed to solve multiattribute group decision making with linguistic Z-numbers. Wherein an extended technique for order preference by similarity to ideal solution is also developed. Finally, a numerical example is shown to illustrate the practicality and effectiveness of the given method.

In this paper, we develop the generalized linguistic weighted logarithm averaging (GLWLA) operator and the generalized linguistic ordered weighted logarithm averaging (GLOWLA) operator in the group decision making under the linguistic surrounding. Then some properties of the families of the GLOWLA operator by different weighting vector are investigated. Furthermore, we present the generalized linguistic ordered weighted hybrid logarithm averaging (GLOWHLA) operator, which extends the GLOWLA operator. We also construct a nonlinear goal programming model to determine GLOWHLA weights from observational linguistic variable values under partial weight information. Finally, a numerical example is given to illustrate the new approach to evaluating university faculty for tenure and promotion, which indicates the feasibility and effectiveness of the new approach.

The aim of this paper is to develop a new compatibility, which is very suitable to deal with group decision making (GDM) problems involving interval multiplicative preference relations, based on the continuous ordered weighted geometric averaging (COWGA) operator. First, we define some concepts of the compatibility degree and the compatibility index for the two interval multiplicative preference relations based on the COWGA operator. Then, we study some desirable properties of the compatibility index and investigate the relationship between each expert's interval multiplicative preference relation and the synthetic interval multiplicative preference relation. The prominent characteristic of the compatibility index based on the COWGA operator is that it can deal with the compatibility of all the arguments in two interval arguments considering the risk attitude of decision maker rather than the compatibility of the two simple points in intervals. Second, in order to determine the experts' weights in the GDM with the interval multiplicative preference relations, we propose an optimal model based on the criterion of minimizing the compatibility index. Finally, we give a numerical example to develop the new approach to GDM with interval multiplicative preference relations.

We develop some new cases of the induced continuous ordered weighted averaging (ICOWA) operator and study their desirable properties, which are very suitable to deal with group decision making (GDM) with interval fuzzy preference relations. First, we present the consensus indicator ICOWA (CI-ICOWA) operator which uses the consensus indicator of the interval fuzzy preference as the order inducing variable in the ICOWA operator. Then the concept of compatibility degree (CD) for two interval fuzzy preference relations is defined based on the continuous ordered weighted averaging (COWA) operator and the compatibility degree ICOWA (CD-ICOWA) operator is proposed which uses the CD as the order inducing variable in the ICOWA operator. Next, we investigate some desirable properties of the CD-ICOWA operator. Additionally, we construct an optimization model to obtain the weights of experts by minimizing the compatibility degree in the GDM. Finally, an illustrative numerical example is used to verify the developed approaches.