This study introduces predictive modeling based on Extreme Gradient Boosting (XGBoost), which utilizes Optuna for hyperparameter optimization and evaluates performance against GridSearchCV and RandomizedSearchCV using a 500-day dataset. To ensure statistical reliability of the findings, a bootstrapping with 1000 iterations is used to calculate 95% confidence intervals for all performance measures. Although GridSearchCV and RandomizedSearchCV achieve consistent performance, their average R-squared test performances are 0.81677 and 0.85538, respectively. In contrast, Optuna outperforms both methods by identifying better regions of optimal parameters, with an average R-squared of 0.94146. Furthermore, the computational efficiency analysis shows that Optuna’s average execution time of 28.31 s is a practical trade-off for running on consumer-grade hardware. Model interpretability is confirmed through Shapley Additive Explanations (SHAP) analysis, which identified the 3-day rolling average as the most important driving factor of demand.
Bearing capacity is a key component affecting the construction of a foundation and land utilization. The range of bearing capacity values can be implemented to determine the land diagnostic methods for sustainably building resilient structures. Various methodologies have been employed in this study to ascertain the bearing capacity of shallow foundations situated on the soil with numerous layers. The ultimate bearing capacity has been calculated using the Indian Standard Code (IS 6403:1981), with the following criteria controlling the process: groundwater level, footing size, embedded depth of the footing, and shear strength parameters. The research has attempted to estimate bearing capacity as a function of these properties using a competent Artificial Neural Network (ANN). The study encompasses the soil data of the Hooghly district of West Bengal. The prediction of bearing capacity in the Hooghly district of West Bengal, based on the soil properties is highlighted in this research.
Image denoising remains a fundamental challenge in low-level vision, particularly in scenarios where noise is nonuniform, structure-dependent, or difficult to model statistically. This paper presents a novel denoising framework that integrates density-based clustering with intuitionistic fuzzy logic to achieve robust detection and suppression of noise pixels. The proposed method first constructs local feature descriptors for every pixel and employs the OPTICS clustering algorithm to identify density anomalies that correspond to potential noise points. For each detected candidate, an intuitionistic fuzzy model computes membership, non-membership, and hesitation degrees based on both local similarity and deviation measures. A pixel-adaptive restoration rule then selectively replaces corrupted pixels using median filtering while preserving uncorrupted structures. Experimental results validate that the proposed OPTICS-fuzzy approach efficiently eliminates noise while maintaining fine image specifics, outperforming conventional clustering-based and fuzzy-based denoising methods across multiple benchmark datasets. The method exhibits strong robustness to varying noise levels and nonGaussian noise distributions, making it suitable for real-world imaging applications.
The research project aims to develop and implement an automated number plate recognition (ANPR) system to enhance traffic control and transportation efficiency. As the number of vehicles on the road continues to grow, the need for an intelligent and automated traffic monitoring system becomes increasingly crucial. The proposed ANPR system leverages machine learning and computer vision algorithms to accurately detect and recognize vehicle license plates in real time. Key functionalities include license plate localization, character segmentation, and alphanumeric character recognition. The system is designed to handle challenges such as partial occlusions, varying angles, and visual distortions. A large, annotated dataset is used for training and evaluation, ensuring high accuracy and adaptability. Successful deployment of the system is expected to enhance law enforcement, strengthen security, and improve traffic flow. Additionally, its applications in smart city initiatives, parking management, and toll collection contribute to a more efficient urban transportation system. The ANPR technology captures images of vehicles passing through checkpoints or surveillance areas and extracts license plate data for rapid identification. With recent advancements significantly improving speed, accuracy, and integration with databases and surveillance networks, ANPR has become a vital tool for modern traffic management.
Liver histopathology plays a critical role in diagnosing and evaluating fibrosis, a condition often assessed through ultrasound imaging. This research proposes an ensemble deep learning model combining ResNet50 and VGG16 architectures, enhanced with attention mechanisms, to effectively classify liver histopathology and fibrosis images. The dataset, comprising grayscale ultrasound images from five classes, was preprocessed by converting to RGB, resizing, and stratified splitting into training, validation, and testing subsets. The model architecture integrates global average pooling and dense layers for classification, while attention blocks amplify salient features, enhancing predictive accuracy. Comprehensive training incorporated data augmentation and early stopping to mitigate overfitting. The model achieved a test accuracy of 97
