The intensity of heavy rainfall, driven by climate change, has significant effects worldwide, including flash flood, droughts, water degradation, landslides and crop damages. To ameliorate these impacts, accurate forecasting is crucial to address the dynamic nature of rainfall for sustainable utilization. But the non-linearity inherited within the rainfall significantly influence the model precision. Artificial Intelligence (AI) models have shown promising results in detecting complex rainfall patterns. This paper proposed a hybrid model using overcomplete rational dilation discrete wavelet transform (ORDWT) integrated with autoregressive integrated moving average (ARIMA) and long-short-term memory (LSTM), constructing ORDWT-ARIMA-LSTM to forecast one-month ahead rainfall. The ORDWT provides multi-scale decomposition and better shift-invariance, while ARIMA with LSTM captures complementary dynamics across ORDWT coefficients, lowering errors. Aiming to extract more representative features, the ORDWT coefficients are investigated, and then sent to the ARIMA-LSTM for prediction. The ORDWT–ARIMA–LSTM achieved highest performance for Melbourne Airport: Root Mean Square Error (RMSE) = 2.9, Mean Absolute Error (MAE) = 1.93, RSE = 0.215, Willmott's Index (WI) = 0.990, Nash–Sutcliffe Index (ENI) = 0.970; Melbourne Botanical Gardens: RMSE = 3.84 MAE = 2.65, RSE = 0.287, WI = 0.710, ENI = 0.962; and Preston Reservoir: RMSE = 3.94 MAE = 2.87 RSE = 0.310, WI = 0.973, ENI = 0.971. The ORDWT–ARIMA–LSTM reduced RMSE by 4.5 % and MAE by 5.3 % on average across stations against comparing models. Results confirmed the efficiency of ORDWT–ARIMA–LSTM in rainfall forecasts, providing valuable support in weather, water management, droughts and floods.

Drone-based imaging is a non-invasive technique that plays a critical role in monitoring plant health, stress levels, and soil moisture. Accurate transformation of RGB images to thermal maps is a challenging task due to various factors, including environmental noise, crop density, and variable lighting. Existing RGB-to-thermal translation approaches typically depend on implicit convolutional representations that inadequately extract directional crop structures, soil background variations, and multi-scale spatial variability, leading to reduce generalization across agricultural environments. To address these limitations, we advance a novel RGB to thermal map transformation approach (DTBQW-2BCLIFT) that integrates Dual-Tree Biquaternion Wavelet Transform (DTBQWT) with a newly designed two-branch convolutional attention layers framework based on Continual Learning with Informed Feature Transfer (2-BCLIFT). RGB drone images are decomposed into 18 feature maps to capture multi-resolution spatial and directional frequency information. Then, DTBQWT coefficients and raw RGB image are fused to 2BCLIFT. Unlike traditional deep learning approaches, the proposed 2-BCLIFT model is specifically optimized for field-level variation and is evaluated on drone imagery collected from four different agricultural fields in Prince Edward Island, Canada, under potato crop cultivation and various environmental conditions. This enables the proposed model to generalize effectively across diverse fields. Several metrics, including RMSE, PSNR, SSIM, and per-pixel accuracy, are employed to evaluate the proposed model against U-Net, Pix2Pix-GAN, Sparse GAN, Thermal-GAN, Modified Pix2Pix, and Cycle GAN. The results demonstrated that the proposed DTBQW-2BCLIFT model provided performance that is well suited for real world agricultural applications and compares favourably with existing state-of-the-art approaches. It can be combined with drones to detect early signs of water stress and nutrient deficiencies through variations in canopy temperature

Federated Learning (FL) is an emerging research area that produces a globally trained model using numerous local users' data and maintains their privacy. Heterogeneous or non-Independent and Identically Distributed ( non-IID) data affect the global model's convergence and, therefore, cause high communication costs. These are because traditional FL approaches often disregard an adaptive regularized objective for the user-side training and utilize conventional arithmetic mean on the locally trained models for the server-side aggregation. To alleviate these issues, we propose a novel FL scheme in this paper. In particular, we propose an adaptive regularization approach to add to the classical objective function of the users' local models during training and a resilient estimation approach to the locally trained models during aggregation. The adaptive regularization approach is derived using the users' local and global performance diversification while the resilient estimation scheme uses a modified geometric mean aggregation over the local models' parameters. We provide consolidated theoretical results and perform extensive experiments on the IID and non-IID settings of MNIST, CIFAR-10, and Shakespeare datasets with various deep networks. The results manifest that our FL scheme outperforms the state-of-the-art approaches in terms of communication speedup, test-set performance, training convergence stability, and resiliency against attacks.

In this paper, we propose a neutrosophic recommender system for medical diagnosis using both neutrosophic similarity measure and neutrosophic clustering to capture the treatment of similar patients at different levels within a concurrent group. The proposed algorithm allows similar patients being treated concurrently in a group. Firstly, the similarities are measured based on the algebraic operations and their theoretic properties. Secondly, a clustering algorithm is used to identify neighbors that are in the same cluster and share common characteristics. Then, a prediction formula using results of both the clustering algorithm and the similarity measures is designed. Experiment indicates the advantages and superiority of the proposal.

Fuzzy sets and intuitionistic fuzzy sets can't handle imprecise, indeterminate, inconsistent, and incomplete information. Neutrosophic sets play an important role to overcome this difficulty. A neutrosophic set has a truth membership function, indeterminate membership function, and a falsehood membership functions that can handle all types of ambiguous information. New type of union and intersection has been proposed in this paper. In this paper, δ -equalities of neutrosophic sets have been introduced. Further, some basic properties of δ - equalities have been discussed. Moreover, these δ - equalities have been applied to set theoretic operations of neutrosophic set such as union, intersection, complement, product, probabilistic sum, bold sum, bold intersection, bounded difference, symmetrical difference, and convex linear sum of min and max. These δ -equalities of neutrosophic sets have been further extended to neutrosophic relations and neutrosophic norms respectively. In this paper, δ -equalities also applied in the composition of neutrosophic relations, Cartesian product and neutrosophic triangular norms. The applications and utilizations of δ -equalities have been presented in this paper. In this regards, δ -equalities have been successfully applied in Fault Tree Analysis and Neutrosophic Reliability (generalization of Profust Reliability).

A complex fuzzy class is characterized by a pure complex fuzzy grade of membership. Pure complex fuzzy classes are paramount in providing rich semantics for cases where the fuzzy data is periodic with a fuzzy period. Often, however, the available data is contaminated by noise, opposing expert opinions, ambiguity, and false information. This opens the door for using intuitionistic fuzzy sets theory: representing the false information via a degree of non-membership. Several researchers have identified the benefits of integrating the two concepts of complex fuzzy sets and intuitionistic fuzzy sets. Nevertheless, complex fuzzy sets allow for only one component of the degree of membership to be fuzzy. In this paper, we introduce the concept of complex intuitionistic fuzzy classes, which are characterized by pure complex intuitionistic fuzzy grade of membership. We define the basic terms and operations on complex intuitionistic fuzzy classes and provide a motivating example of relevant application.

Interval valued bipolar fuzzy weighted neutrosophic set(IVBFWN-set) is a new generalization of fuzzy set, bipolar fuzzy set, neutrosophic set and bipolar neutrosophic set so that it can handle uncertain information more flexibly in the process of decision making. Therefore, in this paper, we propose concept of IVBFWN-set and its operations. Also we give the IVBFWN-set weighted average operator and IVBFWN-set weighted geometric operator to aggregate the IVBFWN-sets, which can be considered as the generalizations of some existing ones under fuzzy, neutrosophic environments and so on. Finally, a decision making algorithm under IVBFWN environment is given based on the given aggregation operators and a real example is used to demonstrate the effectiveness of the method.