Sentiment evolution on social media: An in-depth study using Naive bayes for Twitter sentiment analysis

AIP conference proceedings, Journal Year: 2025, Volume and Issue: 3224, P. 020022 - 020022

Published: Jan. 1, 2025

Language: Английский

---

A Mutual Information Based Approach for Feature Subset Selection and Image Classification

SN Computer Science, Journal Year: 2025, Volume and Issue: 6(4)

Published: April 15, 2025

Language: Английский

---

Computational Intelligence Based Plant Health Management System

Published: March 4, 2025

Language: Английский

---

SAM-LCA: a computationally efficient SAM-based model for tuberculosis detection in chest X-rays

Multimedia Systems, Journal Year: 2025, Volume and Issue: 31(3)

Published: April 26, 2025

Language: Английский

---

Optimized tuberculosis classification system for chest X‐ray images: Fusing hyperparameter tuning with transfer learning approaches

Engineering Reports, Journal Year: 2024, Volume and Issue: 6(11)

Published: April 29, 2024

Abstract Advanced diagnostic methods are necessary for the prompt and reliable identification of tuberculosis (TB), which continues to be a worldwide health problem. Globally, there were projected 10 million new cases in 2021, 9.8 affected adults 0.2 children. About 15% fatalities attributable (1.5 deaths every infections). To create model (TB) using chest X‐ray pictures, we use deep learning approaches this work, namely Convolutional Neural Networks (CNNs) combination transfer hyperparameter tuning. The dataset provides varied selection 3500 normal 700 TB‐infected patients. It consists 4200 photos that obtained from “Tuberculosis Chest Database” on Kaggle. By utilizing benefits trained model, suggested methodological approach incorporates learning. maximize performance adjustment is also used. Using VGG19 pre‐trained neural network, design based concepts architecture makes task‐specific layers, regularization methods, deliberate layer freezing enable sophisticated categorization. Training assessment stages demonstrate encouraging outcomes, with an accuracy almost 98% attained different test dataset. A more thorough examination highlights need caution when interpreting high accuracy, nevertheless, by highlighting possible difficulties.

Language: Английский

---

TSSG-CNN: A Tuberculosis Semantic Segmentation-Guided Model for Detecting and Diagnosis Using the Adaptive Convolutional Neural Network

Diagnostics, Journal Year: 2024, Volume and Issue: 14(11), P. 1174 - 1174

Published: June 1, 2024

Tuberculosis (TB) is an infectious disease caused by Mycobacterium. It primarily impacts the lungs but can also endanger other organs, such as renal system, spine, and brain. When infected individual sneezes, coughs, or speaks, virus spread through air, which contributes to its high contagiousness. The goal enhance detection recognition with X-ray image dataset. This paper proposed a novel approach, named Segmentation-Guided Diagnosis Model (TSSG-CNN) for Detecting Tuberculosis, using combined semantic segmentation adaptive convolutional neural network (CNN) architecture. approach distinguished from most of previously approaches in that it uses combination deep learning model follow-up classification based on CNN layers segment chest images more precisely well improve diagnosis TB. contrasts like ILCM, optimized sequential learning, explainable AI approaches, focus explanations. Moreover, our beneficial simplified procedure feature optimization perspectives Mayfly Algorithm (MA). Other models, including simple CNN, Batch Normalized (BN-CNN), Dense (DCNN), are evaluated this dataset evaluate effectiveness approach. performance TSSG-CNN outperformed all models impressive accuracy 98.75% F1 score 98.70%. evaluation findings demonstrate how works potential further research. results suggest accurate strategy highlight useful technique precise early

Language: Английский

---

Improved Crop Yields and Resource Efficiency in IoT-based Agriculture with Machine Learning

Published: March 14, 2024

Despite popular belief, agricultural research today is more based on hard evidence; exact; precise, and rigorous than ever before. Almost every industry has been disrupted by the spread of IoT-based technologies, including urban planning, healthcare, electricity grid, home, agriculture, frequently referred to as "smart agriculture". Machine learning (ML) IoT data analytics in agriculture can boost crop yields meet rising food demand. These revolutionary developments are upending standard practices giving rise new finest opportunities, but with some drawbacks. Optimal output requires this seeks develop an effective precise system that uses Crop selection choices made using algorithms utilize Internet Things (IoT) sensors ML Therefore, paper, proposed ensemble model machine for prediction which collected from PLX-DAQ tool. There several suggested machines models, "Naive Bayes, Decision Tree, Random Forest Support Vector Machine, K-Nearest Neighbour,". According experimental findings, had greatest accuracy 97.45% predicting early yields. The results will significantly increase dependence relating climate change practices.

Language: Английский

---

Adaptive Gamma Correction with Unsharp Masking and Gaussian filter for Image Contrast Enhancement

Published: March 14, 2024

In this research work image processing is done for direct application of digital images. Image used the enhancement process includes with it. Adaptive histogram equalization an appropriate method improving quality This technique gamma correction. Unsharp masking and Gaussian smoothing filters are also applied to improve visibility as well remove harshness The comparative results different approaches shown in work. proposed superior earlier developed methods by testing on various parameters such MSE, PSNR, MAE, AMBE. suggested strategy maintains brightness types

Language: Английский

---

A Cyber Physical Security for Electrical Vehicles using Deep learning

Published: March 14, 2024

In this paper, Fog-based IoT-enabled Electrical vehicles connected with smart grid systems and propose a reference model that integrates fog computing concepts for electrical grids. Due to its distributed nature, allows capture the real data. capable improve security latency minimization between electric cloud, provide reliability in applications such as vehicles' energy management security. addition, we explain challenges attacks existing solutions. our proposed method CNN+LSTM Algorithms on publicly CICIDS-2017 dataset demonstrated high performance. We achieved an impressive accuracy rate of 99.86%.

Language: Английский

---

An Intelligent Approach to Grape Leaf Disease Diagnosis Through Machine Learning

Published: March 15, 2024

This study underscores the pressing necessity for early and accurate diagnosis of grape leaf diseases, advocating an automated solution that harnesses synergistic power image processing machine learning. Employing methodologies such as Naïve Bayes, Gradient Boosting, Random Forest, primary objective is to proficiently classify leaves either healthy or afflicted by diseases. The model casts its focus on prevalent maladies, including Grapes Healthy Leaf, Grape Black Rot,Grape Measles, Isariopsis leaf. process commences with meticulous preparation images, followed extraction crucial features, which serve foundation application learning techniques. viability crop sector hinges efficacy this innovative approach, has potential significantly curtail spread diseases in fields while enhancing precision disease identification. serves a compelling demonstration formidable effectiveness algorithms detection thus ushering prospects automation improved decision-making within realm agriculture.

Language: Английский

---