Automated Essay Scoring (AES) systems face persistent challenges in delivering accuracy and efficiency in evaluations. This study introduces an approach that combines embeddings generated using RoBERTa with handcrafted linguistic features, leveraging Lightweight XGBoost (LwXGBoost) for enhanced scoring precision. The embeddings capture the contextual and semantic aspects of essay content, while handcrafted features incorporate domain-specific attributes such as grammar errors, readability, and sentence length. This hybrid feature set allows LwXGBoost to handle high-dimensional data and model intricate feature interactions effectively. Our experiments on a diverse AES dataset, consisting of essays from students across various educational levels, yielded a QWK score of 0.941. This result demonstrates the superior scoring accuracy and the model’s robustness against noisy and sparse data. The research underscores the potential for integrating embeddings with traditional handcrafted features to improve automated assessment systems.

On-demand service is the main feature of the 5G network, and Network Function Virtualization (NFV) provides it by virtualizing the existing 5G network infrastructure. NFV crafts various virtual networks on a shared physical network, but one of the core challenges in future 5G networks is to automate the modeling of Virtualized Network Functions (VNFs) and end-to-end Network Service (NS) orchestration with less human interaction. Traditionally, the descriptor of VNF and NS is created manually, which requires expert-level skills. This manual approach has a big threat of human error, which can be avoided by using the Intent-Based Networking (IBN) approach. The IBN approach eliminates the requirement of expertise for designing VNFs and NS by taking users’ intentions as an input. In this paper, the proposed system presents the Intent Management System for VNF modeling and end-to-end NS orchestration for multi-platforms. This system takes the high-level information related to a specific service, configures it accordingly, and converts it into the selected platform. The proposed system is tested using Mobile Central Office Re-architected as Data Center (M-CORD) and Open-Source Management and Orchestration (OSM) orchestrators. The results section shows that the proposed system reduces the effort of the end-user in creating network slices and provides seamless end-to-end service orchestration.

Classification and identification of eye diseases using Optical Coherence Tomography (OCT) has been a challenging task and a trending research area in recent years. Accurate classification and detection of different diseases are crucial for effective care management and improving vision outcomes. Current detection methods fall into two main categories: traditional methods and deep learning-based approaches. Traditional approaches rely on machine learning for feature extraction, while deep learning methods utilize data-driven classification model training. In recent years, Deep Learning (DL) and Machine Learning (ML) algorithms have become essential tools, particularly in medical image classification, and are widely used to classify and identify various diseases. However, due to the high spatial similarities in OCT images, accurate classification remains a challenging task. In this paper, we introduce a novel model called “OCTNet” that integrates a deep learning model combining InceptionV3 with a modified multi-scale attention-based spatial attention block to enhance model performance. OCTNet employs an InceptionV3 backbone with a fusion of dual attention modules to construct the proposed architecture. The InceptionV3 model generates rich features from images, capturing both local and global aspects, which are then enhanced by utilizing the modified multi-scale spatial attention block, resulting in a significantly improved feature map. To evaluate the model’s performance, we utilized two state-of-the-art (SOTA) datasets that include images of normal cases, Choroidal Neovascularization (CNV), Drusen, and Diabetic Macular Edema (DME). Through experimentation and simulation, the proposed OCTNet improves the classification accuracy of the InceptionV3 model by 1.3%, yielding higher accuracy than other SOTA models. We also performed an ablation study to demonstrate the effectiveness of the proposed method. The model achieved an overall average accuracy of 99.50% and 99.65% with two different OCT datasets.

Diabetes is a growing health concern in developing countries, causing considerable mortality rates. While machine learning (ML) approaches have been widely used to improve early detection and treatment, several studies have shown low classification accuracies due to overfitting, underfitting, and data noise. This research employs parallel and sequential ensemble ML approaches paired with feature selection techniques to boost classification accuracy. The Pima India Diabetes Data from the UCI ML Repository served as the dataset. Data preprocessing included cleaning the dataset by replacing missing values with column means and selecting highly correlated features using forward and backward selection methods. The dataset was split into two parts: training (70%), and testing (30%). Python was used for classification in Jupyter Notebook, and there were two design phases. The first phase utilized J48, Classification and Regression Tree (CART), and Decision Stump (DS) to create a random forest model. The second phase employed the same algorithms alongside sequential ensemble methods—XG Boost, AdaBoostM1, and Gradient Boosting—using an average voting algorithm for binary classification. Evaluation revealed that XG Boost, AdaBoostM1, and Gradient Boosting achieved classification accuracies of 100%, with performance metrics including F1 score, MCC, Precision, Recall, AUC-ROC, and AUC-PR all equal to 1.00, indicating reliable predictions of diabetes presence. Researchers and practitioners can leverage the predictive model developed in this work to make quick predictions of diabetes mellitus, which could save many lives.

Millimeter-wave (mmWave) antennas and antenna arrays have gained significant attention due to their pivotal role in emerging wireless communication, sensing, and imaging technologies. With the rapid deployment of 5G and the transition toward 6G networks, the demand for compact, high-gain, and reconfigurable mmWave antennas has intensified. This article highlights recent advancements in mmWave antenna technologies, including hybrid beamforming using phased arrays, dynamic beam-steering enabled by liquid crystal and MEMS-based structures, and high-capacity MIMO architectures. We also examine the integration of metamaterials and metasurfaces for miniaturization and gain enhancement. Applications covered include wearable antennas with low-SAR textile substrates, conformal antennas for UAV-based mmWave relays, and high-resolution radar arrays for autonomous vehicles. The study further analyzes innovative fabrication methods such as inkjet and aerosol jet printing, micromachining, and laser direct structuring, along with advanced materials like Kapton, PDMS, and graphene. Numerical modeling techniques such as full-wave EM simulation and machine learning-based optimization are discussed alongside experimental validation approaches. Beyond communications, we assess mmWave systems for biomedical imaging, security screening, and industrial sensing. Key challenges addressed include efficiency degradation at high frequencies, interference mitigation in dense environments, and system-level integration. Finally, future directions, including AI-driven design automation, intelligent reconfigurable surfaces, and integration with quantum and terahertz technologies, are outlined. This comprehensive synthesis aims to serve as a valuable reference for advancing next-generation mmWave antenna systems.

Home appliances are considered to account for a large portion of smart homes’ energy consumption. This is due to the abundant use of IoT devices. Various home appliances, such as heaters, dishwashers, and vacuum cleaners, are used every day. It is thought that proper control of these home appliances can reduce significant amounts of energy use. For this purpose, optimization techniques focusing mainly on energy reduction are used. Current optimization techniques somewhat reduce energy use but overlook user convenience, which was the main goal of introducing home appliances. Therefore, there is a need for an optimization method that effectively addresses the trade-off between energy saving and user convenience. Current optimization techniques should include weather metrics other than temperature and humidity to effectively optimize the energy cost of controlling the desired indoor setting of a smart home for the user. This research work involves an optimization technique that addresses the trade-off between energy saving and user convenience, including the use of air pressure, dew point, and wind speed. To test the optimization, a hybrid approach utilizing GWO and PSO was modeled. This work involved enabling proactive energy optimization using appliance energy prediction. An LSTM model was designed to test the appliances’ energy predictions. Through predictions and optimized control, smart home appliances could be proactively and effectively controlled. First, we evaluated the RMSE score of the predictive model and found that the proposed model results in low RMSE values. Second, we conducted several simulations and found the proposed optimization results to provide energy cost savings used in appliance control to regulate the desired indoor setting of the smart home. Energy cost reduction goals using the optimization strategies were evaluated for seasonal and monthly patterns of data for result verification. Hence, the proposed work is considered a better candidate solution for proactively optimizing the energy of smart homes.