Road cracks pose a significant challenge to pavement maintenance by affecting structural integrity, jeopardizing traffic safety, and reducing driving comfort. With the limitations of manual inspection methods, such as being labor intensive, time consuming, and prone to human error, automated detection techniques have emerged as efficient and scalable alternatives. This study presents a detailed comparative analysis of state-of-the-art crack detection models, i.e., YOLOv7, VGG-19, ResNet-50, Naive Bayes, and deep convolutional neural networks, evaluating their performance on diverse and complex pavement image datasets. To ensure fairness and consistency, all models were trained and tested under identical conditions. ResNet-50 demonstrated superior performance, achieving the highest accuracy, that is, 99.8% in detecting and segmenting cracks in a variety of pavement scenarios. Its ability to balance precision and robustness makes it a leading solution for automated crack detection.
Heart diseases are a major cause of death worldwide, highlighting the need for early detection. The electrocardiogram (ECG) records the heart’s electrical activity using electrodes. Our research focuses on the ECG data to diagnose heart disorders, particularly arrhythmias. We utilized the MIT-BIH arrhythmia dataset for comparative analysis of various machine learning techniques, including random forest, K-Nearest Neighbor, and Decision Tree, along with deep learning algorithms like Long short-term memory and Convolutional Neural Networks. This required employing various preprocessing methods like filtering and normalization and feature selection techniques such as chi-square and sequential feature selectors to improve the performance of heart disease prediction. Therefore, hybrid machine and deep learning models are proposed, and the results reveal that hybrid models perform better than conventional models.
Automated assessment of tomato crop maturity is vital for improving agricultural productivity and reducing food waste. Traditionally, farmers have relied on visual inspection and manual assessment to predict tomato maturity, which is prone to human error and time-consuming. Computer vision and deep learning automate this process by analysing visual characteristics, enabling data-driven harvest decisions, optimising quality, and reducing waste for sustainable and efficient agriculture. This research demonstrates deep learning models accurately classifying tomato maturity stages using computer vision techniques, utilising a novel dataset of 4,353 tomato images. The Vision Transformer (ViT) model exhibited superior performance in classifying tomatoes into three ripeness categories (immature, mature, and partially mature), achieving a remarkable testing accuracy of 98.67% and the Convolution neural network (CNN) models, including EfficientNetB1, EfficientNetB5, EfficientNetB7, InceptionV3, ResNet50, and VGG16, achieved testing accuracies of 88.52%, 89.84%, 91.16%, 90.94%, 93.15%, and 92.27%, respectively, when tested with unseen data. ViT significantly surpassed the performance of CNN models. This research highlights the potential for deploying ViT in agricultural environments to monitor tomato maturity stages and packaging facilities smartly. Transformer-based systems could substantially reduce food waste and improve producer profits and productivity by optimising fruit harvest time and sorting decisions.
Associative rule mining is a technique for discovering common patterns and correlations in data sets from different databases, including relational, transactional and other types of data repositories, such as relational databases. Different types of patterns exist in data mining such as frequent patterns, extended patterns, regular patterns etc. Many searches have focused on finding the frequent patterns and very little work has been carried out on negative or rare patterns. It has also been observed that only those items which are positively correlated(frequent) are been executed by various algorithms but very less attention is been given to negatively correlated items. Negatively correlated items also called infrequent items are the items which negate with each other. The items which do not satisfy the minimum threshold value generally are always been ignored by many researchers. Mining of Negative association helps in business such as for customer segmentation, in risk management as well as in medical field. So the main aim of writing this paper is to provide a short overview of various research issues involved in finding out positive and negative associations.
The equalization of digital channels is widely recognized as a nonlinear classification problem. In such scenarios, utilizing networks that approximate nonlinear mappings can be highly advantageous. There has also been extensive research on equalizers based on Radial Basis Function Neural Networks (RBFNNs). This study introduces a training methodology centred on the Improved Butterfly Optimization Algorithm (IBOA) for channel equalization using RBFNN. This approach aims to optimize the performance of RBFNN equalizers by leveraging the IBOA algorithm for training. Previous literature primarily approached the equalization problem as an optimization challenge. In contrast, this study addresses it as a classification problem. This training approach exhibits substantial enhancements compared to conventional metaheuristic algorithms.
