Approximate computing is a promising approach for reducing power consumption and design complexity in applications that accuracy is not a crucial factor. Approximate multipliers are commonly used in error-tolerant applications. This paper presents three approximate 4:2 compressors and two approximate multiplier designs, aiming at reducing the area and power consumption, while maintaining acceptable accuracy. The paper seeks to develop approximate compressors that align positive and negative approximations for input patterns that have the same probability. Additionally, the proposed compressors are utilized to construct approximate multipliers for different columns of partial products based on the input probabilities of the two compressors in adjacent columns. The proposed approximate multipliers are synthesized using the 28nm technology. Compared to the exact multiplier, the first proposed multiplier improves power $\times$ delay and area $\times$ power by 91% and 86%, respectively, while the second proposed multiplier improves the two parameters by 90% and 84%, respectively. The performance of the proposed approximate methods was assessed and compared with the existing methods for image multiplication, sharpening, smoothing and edge detection. Also, the performance of the proposed multipliers in the hardware implementation of the neural network was investigated, and the simulation results indicate that the proposed multipliers have appropriate accuracy in these applications.

Advancements in digital technologies have transformed the world by providing more opportunities and possibilities. However, elderly persons have several challenges utilizing modern technology, leading to digital exclusion, which can negatively impact sustainable development. This research attempts to address the current digital exclusion by addressing the challenges older people face considering evolving digital technologies, focusing on economic, social, and environmental sustainability. Three distinct goals are pursued in this study: to perform a detailed literature review to identify gaps in the current understanding of digital exclusion among the elderly, to identify the primary factors affecting digital exclusion in the elderly, and to analyze the patterns and trends in different countries, with a focus on differentiating between High-Income Countries (HICs) and Lower Middle-Income Countries (LMICs). The research strategies used in this study involve a combination of a literature review and a quantitative analysis of the digital exclusion data from five cohorts. This study uses statistical analysis, such as PCA, chi-square test, one-way ANOVA, and two-way ANOVA, to present a complete assessment of the digital issues that older persons experience. The expected results include the identification of factors influencing the digital divide and an enhanced awareness of how digital exclusion varies among different socio-economic and cultural settings. The data used in this study were obtained from five separate cohorts over a five-year period from 2019 to 2023. These cohorts include ELSA (UK), SHARE (Austria, Germany, France, Estonia, Bulgaria, and Romania), LASI (India), MHAS (Mexico), and ELSI (Brazil). It was discovered that the digital exclusion rate differs significantly across HICs and LMICs, with the UK having the fewest (11%) and India having the most (91%) digitally excluded people. It was discovered that three primary factors, including socio-economic status, health-related issues, and age-related limitations, are causing digital exclusion among the elderly, irrespective of the income level of the country. Further analysis showed that the country type has a significant influence on the digital exclusion rates among the elderly, and age group plays an important role in digital exclusion. Additionally, significant variations were observed in the life satisfaction of digitally excluded people within HICs and LMICs. The interaction between country type and digital exclusion also showed a major influence on the health rating. This study has a broad impact since it not only contributes to what we know academically about digital exclusion but also has practical applications for communities. By investigating the barriers that prevent older people from adopting digital technologies, this study will assist in developing better policies and community activities to help them make use of the benefits of the digital era, making societies more equitable and connected. This paper provides detailed insight into intergenerational equity, which is vital for the embedding principles of sustainable development. Furthermore, it makes a strong case for digital inclusion to be part of broader efforts (and policies) for creating sustainable societies.

Fault detection and diagnosis (FDD) systems can reduce high costs and energy consumption. This paper presents a machine learning-based fault detection and diagnosis (FDD) technique for actuators and sensors in a pressurized water reactor (PWR). In the proposed FDD framework, faults are first detected using a shallow neural network. Second, fault diagnosis is performed using 15 different classifiers provided in the MATLAB Classification Learner toolbox, including support vector machine (SVM), K-nearest neighbor (KNN), and ensemble. Several classifiers were found to provide superior classification performance, including medium KNN, cubic KNN, cosine KNN, weighted KNN, fine Gaussian SVM, quadratic SVM, medium Gaussian SVM, coarse Gaussian, bagged trees, and subspace KNN. The accuracy of the FDD approach was demonstrated using a set of simulation results.

Support Vector Machine (SVM) is a supervised machine learning algorithm, which is used for robust and accurate classification. Despite its advantages, its classification speed deteriorates due to its large number of support vectors when dealing with large scale problems and dependency of its performance on its kernel parameter. This paper presents a kernel parameter optimization algorithm for Support Vector Machine (SVM) based on Sliding Mode Control algorithm in a closed-loop manner. The proposed method defines an error equation and a sliding surface, iteratively updates the Radial Basis Function (RBF) kernel parameter or the 2-degree polynomial kernel parameters, forcing SVM training error to converge below a threshold value. Due to the closed-loop nature of the proposed algorithm, key features such as robustness to uncertainty and fast convergence can be obtained. To assess the performance of the proposed technique, ten standard benchmark databases covering a range of applications were used. The proposed method and the state-of-the-art techniques were then used to classify the data. Experimental results show the proposed method is significantly faster and more accurate than the anchor SVM technique and some of the most recent methods. These achievements are due to the closed-loop nature of the proposed algorithm, which significantly has reduced the data dependency of the proposed method.

This paper investigates the use of machine learning techniques on hyperspectral images of pistachios to detect and classify different levels of aflatoxin contamination. Aflatoxins are toxic compounds produced by moulds, posing health risks to consumers. Current detection methods are invasive and contribute to food waste. This paper explores the feasibility of a non-invasive method using hyperspectral imaging and machine learning to classify aflatoxin levels accurately, potentially reducing waste and enhancing food safety. Hyperspectral imaging with machine learning has shown promise in food quality control. The paper evaluates models including Dimensionality Reduction with K-Means Clustering, Residual Networks (ResNets), Variational Autoencoders (VAEs), and Deep Convolutional Generative Adversarial Networks (DCGANs). Using a dataset from Leeds Beckett University with 300 hyperspectral images, covering three aflatoxin levels (160 ppn, and >300 ppn), key wavelengths were identified to indicate contamination presence. Dimensionality Reduction with K-Means achieved 84.38% accuracy, while a ResNet model using the 866.21 nm wavelength reached 96.67%. VAE and DCGAN models, though promising, were constrained by dataset size. The findings highlight the potential for machine learning-based hyperspectral imaging in pistachio quality control, and future research should focus on expanding datasets and refining models for industry application.