An imbalanced dataset often challenges machine learning, particularly classification methods. Underrepresented minority classes can result in biased and inaccurate models. The Synthetic Minority Over-Sampling Techniqu...
详细信息
An imbalanced dataset often challenges machine learning, particularly classification methods. Underrepresented minority classes can result in biased and inaccurate models. The Synthetic Minority Over-Sampling Technique (SMOTE) was developed to address the problem of imbalanced data. Over time, several weaknesses of the SMOTE method have been identified in generating synthetic minority class data, such as overlapping, noise, and small disjuncts. However, these studies generally focus on only one of SMOTE’s weaknesses: noise or overlapping. Therefore, this study addresses both issues simultaneously by tackling noise and overlapping in SMOTE-generated data. This study proposes a combined approach of filtering, clustering, and distance modification to reduce noise and overlapping produced by SMOTE. Filtering removes minority class data (noise) located in majority class regions, with the k-nn method applied for filtering. The use of Noise Reduction (NR), which removes data that is considered noise before applying SMOTE, has a positive impact in overcoming data imbalance. Clustering establishes decision boundaries by partitioning data into clusters, allowing SMOTE with modified distance metrics to generate minority class data within each cluster. This SMOTE clustering and distance modification approach aims to minimize overlap in synthetic minority data that could introduce noise. The proposed method is called “NR-Clustering SMOTE,” which has several stages in balancing data: (1) filtering by removing minority classes close to majority classes (data noise) using the k-nn method;(2) clustering data using K-means aims to establish decision boundaries by partitioning data into several clusters;(3) applying SMOTE oversampling with Manhattan distance within each cluster. Test results indicate that the proposed NR-Clustering SMOTE method achieves the best performance across all evaluation metrics for classification methods such as Random Forest, SVM, and Naїve Bayes, compared t
This study presents a novel ultra-high step-up (UHSU) DC-DC topology tailored for applications in DC microgrids. The proposed configuration utilizes a quadraticbased topology, achieving a remarkably high voltage gain ...
详细信息
This research allows the secure surveillance approach for the Internet of Things (IoT) methodology to be developed by integrating wireless signalling and image encryption strategy. Since the Cloud Service Telco (CST) ...
详细信息
Particles in the atmosphere, such as dust and smoke, can cause visual clarity problems in both images and videos. Haze is the result of the interaction between airborne particles and light, which is scattered and atte...
详细信息
Particles in the atmosphere, such as dust and smoke, can cause visual clarity problems in both images and videos. Haze is the result of the interaction between airborne particles and light, which is scattered and attenuated. Hazy media present difficulties in a variety of applications due to the reduced contrast and loss of essential information. In response, dehazing techniques have been introduced to bring hazy videos and images back to clarity. Here, we provide a novel technique for eliminating haze. It comprises preprocessing steps before dehazing. Preprocessing is applied to hazy images through homomorphic processing and Contrast Limited Adaptive Histogram Equalization (CLAHE). We present a dehazing technique referred to as the pre-trained Feature Fusion Attention Network (FFA-Net) that directly lets dehazed images be restored from hazy or preprocessed hazy inputs without requiring the determination of atmospheric factors, such as air light and transmission maps. The FFA-Net architecture incorporates a Feature Attention (FA) method to do this task. We assess the proposed technique in a variety of circumstances, including visible frames, Near-Infrared (NIR) frames, and real-world hazy images. Evaluation criteria like entropy, correlation, and Peak Signal-to-Noise Ratio (PSNR) are used to compare the quality of dehazed frames or images to their hazy counterparts. Furthermore, histogram analysis and spectral entropy are adopted to determine the effectiveness of the proposed technique in comparison to existing dehazing techniques. Comparative results are presented for both real-world and simulated environments. The benefits of the proposed technique are demonstrated by a comparison of the results obtained from the standalone pre-trained FFA-Net and the proposed comprehensive methodology. Moreover, a thorough assessment is carried out for comparing the effectiveness of the proposed FFA-Net technique to those of some current dehazing techniques on real hazy images. T
In this paper, we propose a cloud-based virtual commissioning Web platform for IEC 61499 compliant distributed automation systems that support Create-Read-Update-Delete operations for SoftPLC containers from various v...
详细信息
This paper presents a novel framework for testing automation applications compliant with the international standard IEC 61499 and including process simulation. The framework enables automation programs to be run in te...
详细信息
The efficient distribution of function blocks across programmable logic controllers (PLCs) is critical to improving the performance of distributed automation systems. This paper presents a novel randomized greedy opti...
详细信息
As the modern bulk power system (BPS) expands and electricity users increase, an ever-growing amount of power data needs to be processed. The recommender system (RS), as an artificial intelligence (AI) algorithm, leve...
详细信息
This paper proposes an extension to IEC 61499 architecture to better support flexibility and reconfigurability of automation systems while reducing communication overhead. The extension concerns adapter connections be...
详细信息
This analysis was made on the structure of the permanent magnets located on the rotor exterior of the BLDC electric scooter motor, three different case studies were carried out to allow us a view of the change in the ...
详细信息
暂无评论