Abstract The heterogeneous layers of attributed multiplex networks (AMNs) encode structurally distinct and spectrally diverse relational patterns. However, traditional Graph Neural Network models often overlook this frequency diversity, applying uniform filtering and layer fusion strategies that fail to capture cross-layer complementarity. We propose AMN-ARMA, a frequency-adaptive representation learning framework that models each layer as operating in a distinct spectral regime. It combines adaptive ARMA (AutoRegressive Moving Average) filters tailored to layer-specific frequency responses with multi-band, entropy-regularized attention to ensure balanced and interpretable integration. We further introduce a training objective that encourages frequency specialization across layers while maintaining both intra- and inter-layer consistency. Experiments on three real-world AMN datasets demonstrate that AMN-ARMA achieves up to 1.8% improvement in cross-layer link prediction and 2.6% gain in clustering quality, while maintaining robustness to structural perturbations and spectral noise. Spectral analyses reveal that AMN layers indeed operate in complementary frequency bands, and that the learned filters adapt accordingly–validating the framework design motivation and theoretical underpinnings.

Sentiment models that perform well on short tweets often struggle with longer reviews, while review-optimized systems may miss sparse cues in noisy, slang-filled tweets. Tweets typically contain few explicit sentiment indicators and sudden polarity changes; reviews spread evidence across multiple sentences, requiring long-range modeling without issues like unstable optimization or vanishing gradients. The high dimensionality and sparsity increase computational demands and decrease cross-domain robustness. We introduce DAFN, a polarity-detection framework designed to reliably handle both short and long texts. DAFN combines two embeddings: a semantic embedding that captures meaning and a sentiment-similarity embedding that intensifies sentiment signals when each token is important. The combined sequence is processed by parallel bidirectional BiGRU and BiLSTM modules to leverage complementary temporal dynamics. Our main innovation is the Attention Interaction module, which explicitly links the branches so that attention from one refines the other, enhancing focus on sentiment-rich segments and providing gradient paths for stable training. DAFN further employs branch-specific multi-scale temporal convolutions with combined pooling to identify phrase-level patterns and produce a concise representation, reducing dimensionality while preserving key cues. Lastly, multi-level fusion integrates information at early (embeddings), middle (attention interaction), and late (shortcut-based feature concatenation) stages, strengthening information flow and avoiding the fragility of single-stage fusion. Experiments on standard tweet and review datasets demonstrate consistent improvements in F1 and AUC scores, better separating classes in short, noisy texts, while remaining robust on longer documents.

Accurate flight delay prediction is fundamental to establishing an efficient airline business. It is considered one of the most critical intelligent aviation systems components. Recently, flight delay has been a significant cause that deprives airlines of good performance. Hence, airlines must accurately forecast flight delays and comprehend their sources to have excellent passenger experiences, increase income and minimise unwanted revenue loss. In this paper, we developed a novel approach that is an optimisation-driven deep learning model for predicting flight delays by extending a state-of-the-art method, DeepONet. We utilise the Box-Cox transformation for data conversion with a minimal error rate. Also, we employed a deep residual network for the feature fusion before training our model. Furthermore, this research uses flight on-time data for flight delay prediction. To validate our proposed model, we conducted a numerical study using the US Bureau of Transportation of Statistics. Also, we predict the flight delay by selecting the optimum weights using the novel DeepONet with the Gradient Mayfly Optimisation Algorithm (GMOA). Our experiment results show that the proposed GMOA-based DeepONet outperformed the existing methods with a Root Mean Square Error of 0.0765, Mean Square Error of 0.0058, Mean Absolute Error of 0.0049 and Mean Absolute Percent Error of 0.0043, respectively. When we apply 4-fold cross-validation, the proposed GMOA-based DeepONet outperformed the existing methods with minimal standard error. These results also show the importance of optimisation algorithms in deciding the optimal weight to improve the model performance. The efficacy of our proposed approach in predicting flight delays with minimal errors well define from all the evaluation metrics. Also, utilising the prediction outcome of our robust model to release information about the delayed flight in advance from the aviation decision systems can effectively alleviate the passengers’ nervousness.

Abstract The importance of robust flight delay prediction has recently increased in the air transportation industry. This industry seeks alternative methods and technologies for more robust flight delay prediction because of its significance for all stakeholders. The most affected are airlines that suffer from monetary and passenger loyalty losses. Several studies have attempted to analysed and solve flight delay prediction problems using machine learning methods. This research proposes a novel alternative method, namely social ski driver conditional autoregressive-based (SSDCA-based) deep learning. Our proposed method combines the Social Ski Driver algorithm with Conditional Autoregressive Value at Risk by Regression Quantiles. We consider the most relevant instances from the training dataset, which are the delayed flights. We applied data transformation to stabilise the data variance using Yeo-Johnson. We then perform the training and testing of our data using deep recurrent neural network (DRNN) and SSDCA-based algorithms. The SSDCA-based optimisation algorithm helped us choose the right network architecture with better accuracy and less error than the existing literature. The results of our proposed SSDCA-based method and existing benchmark methods were compared. The efficiency and computational time of our proposed method are compared against the existing benchmark methods. The SSDCA-based DRNN provides a more accurate flight delay prediction with 0.9361 and 0.9252 accuracy rates on both dataset-1 and dataset-2, respectively. To show the reliability of our method, we compared it with other meta-heuristic approaches. The result is that the SSDCA-based DRNN outperformed all existing benchmark methods tested in our experiment.

The Unified Parallel C (UPC) language from the Partitioned Global Address Space (PGAS) family unifies the advantages of shared and local memory spaces and offers a relatively straightforward code parallelisation with the Central Processing Unit (CPU). In contrast, the Computer Unified Device Architecture (CUDA) development kit gives a tool to make use of the Graphics Processing Unit (GPU). We provide a detailed comparison between these novel techniques through the parallelisation of a two-dimensional lattice Boltzmann method based fluid flow solver. Our comparison between the CUDA and UPC parallelisation takes into account the required conceptual effort, the performance gain, and the limitations of the approaches from the application oriented developers’ point of view. We demonstrated that UPC led to competitive efficiency with the local memory implementation. However, the performance of the shared memory code fell behind our expectations, and we concluded that the investigated UPC compilers could not efficiently treat the shared memory space. The CUDA implementation proved to be more complex compared to the UPC approach mainly because of the complicated memory structure of the graphics card which also makes GPUs suitable for the parallelisation of the lattice Boltzmann method.

Computational complexity theory inspires a grid-free abstraction of the Navier-Stokes equations in Fortran 95/2003. A novel complexity analysis estimates that structured programming time grows at least quadratically with the number of program lines. Further analysis demonstrates how an object-oriented strategy focused on mathematical objects renders the quadratic estimate scale-invariant, so the time required for the limiting factor in program development (debugging) no longer grows as the code grows. Compared to the coordinate-free C++ programming of Grant et al. [2000], grid-free Fortran programming eliminates a layer of procedure calls, eliminates a related need for the C++ template construct, and offers a shorter migration path for Fortran programmers. The grid-free strategy is demonstrated by constructing a physical-space driver for a Fourier-space Navier-Stokes solver. Separating the expression of the continuous mathematical model from the discrete numerics clarifies issues that are otherwise easily conflated. A run-time profile suggests that grid-free design substantially reduces the fraction of the procedures that significantly impact runtime, freeing more code to be structured in ways that reduce development time. Applying Amdahl's law to the total solution time (development time plus run time) leads to a strategy that negligibly impacts development time but achieves 58% of the maximum possible speedup.

Geometric Multigrid methods have gained widespread acceptance for solving large systems of linear equations, especially for structured grids. One of the challenges in successfully extending these methods to unstructured grids is the problem of generating an appropriate set of coarse grids. The focus of this paper is the development of robust algorithms, both serial and parallel, for generating a sequence of coarse grids from the original unstructured grid. Our algorithms treat the problem of coarse grid construction as an optimization problem that tries to optimize the overall quality of the resulting fused elements. We solve this problem using the multilevel paradigm that has been very successful in solving the related grid/graph partitioning problem. The parallel formulation of our algorithm incurs a very small communication overhead, achieves high degree of concurrency, and maintains the high quality of the coarse grids obtained by the serial algorithm.