Modern Standard Arabic's (MSA) rich and agglutinative morphology, complexity of clitics, lack of diacritical marks, and resulting morphosyntactic ambiguities make POS-tagging difficult. In order to test for ambiguous forms and stylistic variants, this study assesses the robustness of AraBERT, MARBERT, and CAMeL Tools on an enriched corpus of 8,147 sentences (293,199 tokens), which includes 400 synthetic sentences and 200 literary sentences. The findings demonstrate the limitations of the models when exposed to a range of text types with strong performances on journalistic texts (F1 Macro ~83.5%) and a decline on synthetic (~66%) and literary (~55%) data. To enhance the investigation of ambiguity, we also suggest the PL-Score, a supplementary measure that assesses errors based on their linguistic plausibility (e.g., NOUN→ADJ). These findings underscore the necessity for varied corpora and robust hybrid methodologies combining deep learning and human linguistic knowledge to significantly enhance POS-tagging with implications for machine translation and literary text analysis.

Condition-based maintenance (CbM) is a successful cost-cutting approach that is used to reduce production losses by avoiding breakdowns via effective continuous monitoring of the production equipment. Machine Learning (ML) algorithms can be used for this purpose in terms of predicting defects and diagnostics due to the huge amount of data generated by the incorporation of advance analog and digital technologies in manufacturing industries. To improve the prediction accuracy of the CbM, this study proposed a new framework using various ML algorithms. The new framework has two prediction stages: one to classify whether the status of the machine is on an operation mode or a failure mode (binary classification), and a second one to classify the type of the failure (multi-class classification). A comparative study using a public datasets is used to evaluate the proposed ML algorithms. To address the problem of data unbalance, a new modified data was introduced using the RandomOverSampler method. The proposed ML models were implemented on both the original public dataset and its modified version to perform binary and multi-class classification tasks. The hybrid XGBoost-DT prediction model achieves the best and most robust classification and prediction accuracy.

Mechatronic systems, including robotic arms and CNC machines, can experience operational failures, increased downtime, and financial losses due to software flaws. Through the analysis of sensor data, system logs, and operational metrics, this study suggests a hybrid machine learning (ML) framework for anomaly identification and fault prediction in mechatronic systems. Three main modules make up the framework: (1) a feature extraction module that uses time-series and statistical analysis to extract important indicators; (2) an anomaly detection module that uses Autoencoders and Isolation Forest to find anomalous patterns; and (3) a fault prediction module that uses a Random Forest and Multi-Layer Perceptron (MLP) ensemble for accurate fault classification. Real-world industrial datasets and benchmark datasets, including the NASA Bearing Dataset and PHM Data Challenge datasets, are used to assess the suggested approach. According to experimental results, the Fault Prediction Module achieves 96.3% accuracy with an AUC of 0.98, while the Anomaly Detection Module achieves 94.5% accuracy. The framework improves predictive maintenance techniques, lowers false alarms, and effectively detects anomalies caused by software. This study demonstrates how ML-driven fault detection can enhance industrial mechatronic systems' dependability, minimize downtime, and optimize maintenance schedules. For additional improvement, future research will concentrate on edge computing integration, real-time deployment, and adaptive learning models.

Serverless architecture has emerged as a transformative cloud computing model, offering SMEs scalable, cost-efficient alternatives to traditional infrastructure. By 2025, its adoption has accelerated, with the market projected to grow at a 22.2% CAGR, driven by benefits like pay-per-use pricing and automatic scaling. However, challenges such as cold starts, vendor lock-in, and security risks persist, particularly for SMEs with limited IT resources. This study investigates how serverless architectures can be optimized for SMEs, focusing on cost efficiency, scalability, and integration challenges. It aims to identify workload-specific trade-offs and provide actionable adoption frameworks. A mixed-methods approach was employed: quantitative benchmarking of AWS Lambda/Azure Functions versus traditional VMs under SME workloads (bursty traffic, batch processing, IoT streams). Qualitative case studies of SME implementations (e.g., e-commerce, healthcare) were used to analyze operational and security barriers. The goal was to design science research to develop a hybrid architecture decision matrix trial. Potential cost savings of 47-62% for variable workloads, but VM has superiority (12-18% cheaper) for sustained loads. There is 10x faster scaling with serverless, though cold starts affected 5.7% of invocations (1.4s latency). Hybrid architectures reduced TCO by 33% for mixed workloads. There are several security gaps including: 35% of SMEs faced IAM misconfigurations and serverless architectures are highly viable for SMEs with unpredictable workloads but require workload profiling and hybrid designs to mitigate limitations. Future research should explore AI-driven autoscaling and multi-cloud portability to address vendor lock-in.

This paper introduces a new method known as “Enhanced Swarm Distributed Energy-Efficient Clustering” (ES-DEEC). It was designed to resolve critical problems such as energy depletion, network instability, and unreliable data transmission in complex WSN environments. This approach utilizes a hybrid Grey Wolf Optimizer-Particle Swarm Optimization (GWO-PSO) technique with congestion-aware routing and dynamic Cluster-Head (CH) selection to achieve balanced energy consumption (EC) and improved stability of the network. The findings demonstrate superior performance by achieving a lower End-To-End Delay (E2D) of 8ms, 80% Packet Delivery Ratio (PDR), EC by 28 Joules, network Throughput (Th) of 89% and a Network LifeTime (NLT) of 90%. These results are also compared to that of conventional protocols such as Energy-Efficient Cluster-based Routing (ECRR), DEEC, Internet-of-Things DEEC (IoT-DEEC), and Routing based on Tree and Geographic (RTG) to validate the proposed ES-DEEC protocol as a scalable and reliable solution for applications requiring long-term energy-efficient operations, i.e. smart cities and industrial IoT. The results show that the newly presented ES-DEEC protocol performed better than any of the currently selected existing standard protocols. According to these results, it is a superior option for WSNs of the future.

Wireless Sensor Networks (WSNs) are critical in modern applications—from environmental monitoring and innovative healthcare to industrial automation. However, their decentralized and resource-constrained nature makes them inherently vulnerable to sophisticated security threats. Traditional perimeter-based security frameworks, which assume internal nodes are trustworthy, have proven insufficient against insider attacks, and advanced persistent threats. Zero Trust (ZT), with its “never trust, always verify” philosophy, offers a paradigm shift by distrusting every entity by default and enforcing continuous verification. In this paper, a comprehensive Zero Trust framework tailored for WSNs is proposed, addressing key challenges such as dynamic trust establishment, constant authentication of sensor nodes, and energy-efficient security mechanisms suitable for constrained devices. Simulation results demonstrate improved resilience against various attack types (e.g., Sybil, blackhole, wormhole, denial of service), reduced false positives in anomaly detection via machine learning, and optimized energy consumption through trust-based clustering. Further discuss challenges in scaling Zero Trust in heterogeneous WSN environments and outline future directions, including integration with emerging security paradigms, machine learning-driven adaptive security, blockchain-based trust management, and energy-aware trust models. This work contributes to advancing secure communication in WSNs using a Zero Trust approach, highlighting its advantages and remaining challenges in real-world deployments.