Accurate forecasting of reference evapotranspiration (ETo) is crucial for sustainable water resource management and precision agriculture. The present study evaluates three ETo prediction methods: Random Forest (RF), Cartesian Genetic Programming (CGP), and Convolutional Neural Network-Graphics Processing Unit (CNN-GPU) across time intervals of 1 to 364 days. Using dispersion analysis (scatter/violin plots) and accuracy metrics (RMSE, MAE, R^2, SI), it was seen that the RF and CNN-GPU models consistently outperform CGP, particularly at extended horizons. At 364 days, CNN-GPU achieved the highest accuracy (RMSE: 0.678 mm/day, R^2: 0.874), while RF maintained robust performance (RMSE: 0.683 mm... More >

The growing prominence of prompt engineering as a means of controlling large language models has given rise to a diverse set of methods, ranging from handcrafted templates to embedding-level tuning. Yet, as prompts increasingly serve not merely as input scaffolds but as adaptive interfaces between users and models, the question of how to systematically optimize them remains unresolved. Reinforcement learning, with its capacity for sequential decision-making and reward-driven adaptation, has been proposed as a possible framework for discovering effective prompting strategies. This survey explores the emerging intersection of RL and prompt engineering, organizing existing research along three... More >

The rapid advancement of generative AI has created a critical gap between technological innovation and responsibility frameworks. This article examines the comprehensive challenges posed by AI systems that can autonomously generate content and make decisions affecting crucial social domains. We analyze the failure of traditional accountability mechanisms in addressing AI's emergent behaviors and ``black box'' characteristics, and propose a multi-dimensional approach to responsibility allocation. The analysis covers five key areas: the primary responsibilities of technology developers throughout the AI lifecycle, the necessary paradigm shifts in legal frameworks including new concepts of algo... More >

The rapid expansion of edge computing and Internet of Things (IoT) ecosystems has introduced new cybersecurity challenges, particularly in decentralized, resource-constrained environments where traditional security models often fall short. This paper proposes an immune-inspired artificial intelligence framework (I3AI) that draws on core principles of biological immune systems including self-organization, local learning, and immune memory to enable adaptive, privacy-preserving defense mechanisms across distributed edge nodes. The architecture incorporates federated learning to maintain a decentralized threat intelligence network while ensuring data privacy and minimal communication overhead.... More >

This study systematically reviews the current status and recent advances in intelligent depression detection, aiming to provide insights for applying artificial intelligence in mental health. Using a systematic review approach, we analyze detection methods based on multiple data types including voice, facial expressions, body signals, and social media texts, while examining how algorithms have evolved from traditional machine learning to deep learning. Results show that AI technology has clear benefits in improving detection accuracy, reducing costs, and enabling early warning systems. Current research still faces important challenges in data collection, technical reliability, clinical use,... More >

In the medical field, efficient and accurate classification of sequential and structured data is crucially important and useful for early diagnosis and treatment. Traditional machine learning models struggle with the complexity and nonlinearity of dynamic datasets, whereas deep learning models, despite their effectiveness, require extensive resources and lack transparency. This paper proposes a novel lightweight ensemble framework integrating a parameterized SoftMax function with a non-parametric Random Forest method through a soft voting mechanism, supported by the Nonlinear AutoRegressive eXogenous (NARX) model and optimized using a forward orthogonal search and selection (FOSS) algorithm... More >

Topic modeling is a widely used unsupervised natural language processing (NLP) technique aimed at discovering latent themes within documents. Since traditional methods fall short in capturing contextual meaning, approaches based on large language models (LLMs)—such as BERTopic—hold the potential to generate more meaningful and diverse topics. However, systematic comparative studies of these models, especially in domains requiring high accuracy and interpretability such as healthcare, remain limited. This study compares ten different LLMs (GPT, Claude, Gemini, LLaMA, Qwen, Phi, Zephyr, DeepSeek, NVIDIA-LLaMA, Gemma) using a dataset of 9,320 medical article abstracts. Each model was tasked... More >

Federated Learning (FL) which eliminates the centralized data storage requirement by facilitating model training on diverse edge devices is now a promising paradigm for decentralized machine learning (ML). Applications involving privacy-preserving Artificial Intelligence (AI), including wearable technology, IoT networks, and smart healthcare appliances, can particularly benefit from this solution in embedded systems. By using on-device local data from devices such as sensors, embedded controllers, and smartphones, FL keeps confidential information local, minimizing the data transfer cost and privacy risks. Potentiality, challenges, and key applications of FL integration with embedded systems... More >

This article [1] has been retracted by ICCK following an investigation conducted by the publisher. After publication, it was brought to the journal’s attention that some of the listed authors were unaware of the submission and had not provided their consent to be included as co-authors. In accordance with the COPE guidelines [2], the publisher initiated a formal investigation. It was confirmed that the author Shuyan Liu (School of Information Science and Technology, Yunnan University, Yunnan 650000, China) was unaware of the submission, did not contribute to the research or writing of the manuscript, and did not approve the final version for publication. As a result, the article is b... More >

The integration of Artificial Intelligence (AI) with cloud computing has emerged as a pivotal strategy for enterprises seeking scalable and intelligent modernization. This paper explores how cloud-based AI solutions are transforming enterprise ecosystems by offering highly scalable, flexible, and cost-effective platforms for deploying intelligent applications. We examine the convergence of AI-as-a-Service (AIaaS), cloud-native architectures, and data-driven decision-making, and how these capabilities collectively drive operational efficiency, customer engagement, and innovation—particularly within sectors such as healthcare, finance, and manufacturing. The study investigates key enablers i... More >

Abnormal fluctuations in financial markets may signal significant risks or market manipulation, so efficient time series anomaly detection methods are crucial for risk management. However, traditional statistical methods (e.g., ARIMA, GARCH) are difficult to adapt to the nonlinear and multi-scale characteristics of financial data, while single deep learning models (e.g., LSTM, Transformer) have limitations in capturing long-term trends and short-term fluctuations. In this paper, we propose WaveLST-Trans, a financial time series anomaly detection model based on the combination of wavelet transform (WT), LSTM and Transformer. The model first uses wavelet transform to perform multi-scale decomp... More >

Predicting personality traits automatically has emerged as a challenging problem in computer vision. This paper introduces an innovative multimodal feature learning framework for personality analysis in short video clips. For visual processing, we construct a facial graph and design a Geo-based two-stream network incorporating an attention mechanism, leveraging both Graph Convolutional Networks (GCN) and Convolutional Neural Networks (CNN) to capture static facial expressions. Additionally, ResNet18 and VGGFace networks are employed to extract global scene and facial appearance features at the frame level. To capture dynamic temporal information, we integrate a BiGRU with a temporal attentio... More >

Public health surveillance is crucial for early disease detection, outbreak prediction, and epidemic response. However, traditional surveillance systems primarily rely on structured clinical data, limiting their capacity to capture emerging health threats from diverse and unstructured sources. This study explores the integration of Natural Language Processing (NLP) and Artificial Intelligence (AI) to automate disease surveillance by analyzing unstructured data, including electronic health records (EHRs), social media posts, news reports, and online health forums. Leveraging state-of-the-art NLP techniques—such as transformer-based language models, named entity recognition (NER), sentiment... More >

This editorial explores the transformative role of generative artificial intelligence (AI) in augmenting human creativity and catalyzing cognitive evolution. Tracing its historical lineage from symbolic AI to transformer-based architectures, this editorial argues that generative AI is not merely a computational tool but a cognitive partner that reshapes our understanding of creativity, perception, and epistemology. The phenomenon of AI hallucination—often dismissed as error—is reframed as a window into the dynamics of both artificial and human cognition. Through technical and philosophical analysis, the paper discusses generative AI’s impact on fields ranging from art and architecture... More >

The scalability of modern AI is fundamentally limited by the availability of labeled data. While supervised learning achieves remarkable performance, it relies on large annotated datasets, which are expensive and time-consuming to acquire. This work explores self-supervised learning (SSL) as a promising solution to this challenge, enabling AI to scale effectively in data-scarce scenarios. This study demonstrates the effectiveness of the proposed SSL framework using the EuroSAT dataset, a benchmark for land cover classification where labeled data is limited and costly. The proposed approach integrates contrastive learning with multi-spectral augmentations, such as spectral jittering and band... More >

Interpreting NMR spectra to accurately predict molecular structures remains a significant challenge in chemistry due to the complexity of spectral data and the need for precise structural elucidation. This study introduces NMRGen, a generative modeling framework that predicts molecular structures from NMR spectra and molecular formulas. The framework combines a SMILES autoencoder (GRU-based encoder-decoder) and an NMR encoder (CNN and DNN layers) to map spectral data to molecular representations. The SMILES autoencoder compresses and reconstructs SMILES strings, while the NMR encoder processes NMR spectra to generate latent vectors aligned with those from the SMILES encoder. Experiments were... More >

With the increasing global focus on renewable energy and the growing proportion of renewable power in the energy mix, accurate forecasting of renewable power demand has become crucial. This study addresses this challenge by proposing a multimodal information fusion approach that integrates time series data and textual data to leverage complementary information from heterogeneous sources. We develop a hybrid predictive model combining CNN and Bi-GRU architectures. First, time series data (e.g., historical power generation) and textual data (e.g., policy documents) are preprocessed through normalization and tokenization. Next, CNNs extract spatial features from both data modalities, which are... More >

Recently, the primary reason for blindness in adults has been diabetic retinopathy (DR) disease. Therefore, there is an increasing demand for a real-time efficient classification and detection system for diabetic retinopathy (DR) to overcome fast-growing disease (DR). We introduced a novel deep hybrid model for auto-mated diabetic retinopathy (DR) disease recognition and classification. Our model leverages the power of CNN architectures: Inception V3 and VGG16 models by combining their strengths to cater to exact requirements. VGG16 model efficiently captures fine features and wide-ranging features such as textures and edges, crucial for classifying initial signs of DR. Similarly, Inception... More >

This paper proposes an improved video action recognition method, primarily consisting of three key components. Firstly, in the data preprocessing stage, we developed multi-temporal scale video frame extraction and multi-spatial scale video cropping techniques to enhance content information and standardize input formats. Secondly, we propose a lightweight Inception-3D networks (LI3D) network structure for spatio-temporal feature extraction and design a soft-association feature aggregation module to improve the recognition accuracy of key actions in videos. Lastly, we employ a bidirectional LSTM network to contextualize the feature sequences extracted by LI3D, enhancing the representation capa... More >

In the era of rapid technological advancement, the demand for sophisticated Multi-Object Tracking (MOT) systems in applications such as intelligent surveillance and autonomous navigation has become increasingly critical. However, existing models often struggle with accuracy and efficiency in densely populated or dynamically complex environments. Addressing these challenges, we introduce a novel deep learning-based MOT model that incorporates the latest CT-DETR detection technology and an advanced ReID module for improved pedestrian tracking. Experimental results demonstrate the model's superior performance in accurately identifying and tracking multiple targets across varied scenarios, signi... More >

This paper proposes an improved object detection algorithm based on a dynamically deformable convolutional network (D-DCN), aiming to solve the multi-scale and variability challenges in object detection tasks. First, we review traditional methods in the field of object detection and introduce the current research status of improved methods based on multi-scale and variability convolutional neural networks. Then, we introduce in detail our proposed improved algorithms, including an improved feature pyramid network and a dynamically deformable network. In the improved feature pyramid network, we introduce a multi-scale feature fusion mechanism to better capture target information at different... More >

In this paper, a novel fast object detection framework is introduced, designed to meet the needs of real-time applications such as autonomous driving and robot navigation. Traditional processing methods often trade off between accuracy and processing speed. To address this issue, a hybrid data representation method is proposed that combines the computational efficiency of voxelization with the detail capture capability of direct data processing to optimize overall performance. The detection framework comprises two main components: a Rapid Region Proposal Network (RPN) and a Refinement Detection Network (RefinerNet). The RPN is used to generate high-quality candidate regions, while the Refine... More >

With the rapid development of autonomous driving technology, the demand for real-time and efficient object detection systems has been increasing to ensure vehicles can accurately perceive and respond to the surrounding environment. Traditional object detection models often suffer from issues such as large parameter sizes and high computational resource consumption, limiting their applicability on edge devices. To address this issue, we propose a lightweight object detection model called YOLOv8-Lite, based on the YOLOv8 framework, and improved through various enhancements including the adoption of the FastDet structure, TFPN pyramid structure, and CBAM attention mechanism. These improvement... More >