检索结果-内蒙古大学图书馆

3rd IEEE Delhi Section Flagship Conference, DELCON 2024

作者： Chetan, Gudditi Lavanya, Kotha Krishna, S.B.S.S.S. Vamsi Lakshmi, Dasari Venkata School of Computer Science and Engineering VIT-AP University Vijayawada India

ISBN: (纸本)9798331518592

This research presents a method to enhance plant disease identification using Convolutional Neural Networks (CNNs) by leveraging a diverse dataset sourced from Kaggle, which contains 20,000 high-resolution images of healthy and diseased plants. The proposed model evaluates multiple CNN architectures, including ResNet50V2, MobileNetV2, InceptionV3, NASNetMobile, and an ensemble model combining MobileNetV2 and InceptionV3. The models were rigorously trained and assessed on their ability to accurately classify plant diseases. The results showed that ResNet50V2 achieved the highest accuracy at 92.26%, followed by the custom CNN at 88.22%, the ensemble model at 88.08%, MobileNetV2 at 87.93%, InceptionV3 at 82.26%, and NASNetMobile at 82.07%. © 2024 IEEE.

关键词： Plant diseases

来源：评论

学校读者我要写书评

暂无评论

Actionable Intelligence for Sales Call Recordings 3

Actionable Intelligence for Sales Call Recordings

引用

3rd IEEE Delhi Section Flagship Conference, DELCON 2024

作者： Neha, Palanati Chetan, Gudditi Garimella, Srinaini Sreejagadeesh, Malla Naseeba, Beebi School of Computer Science and Engineering VIT-AP University Vijayawada India

ISBN: (纸本)9798331518592

This research highlights the challenges associated with assessing large volumes of recorded discussions, including issues with data privacy and legal compliance. This research main objective is to increase customer happiness and sales performance by using data-driven insights from sales call records. The dataset consists of sales call recordings transcribed into text, containing customer interactions, questions, objections, and competitor mentions. This work utilizes natural language processing (NLP) techniques such as Named Entity Recognition (NER), sentiment analysis (using VADER), and keyword extraction. The results revealed key insights, such as customer pain points, sentiment trends, competitor mentions, and common objections. The negative statements of text is around 2.7%, neutral statements of text reflects 69.7%, positive statements of text reflects 27.6% and compound statements of text reflects 0.9999%. © 2024 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

Extractive Multi-Document Summarization Using Transformer-Based Neural Networks 4

Extractive Multi-Document Summarization Using Transformer-Ba...

引用

4th International Conference on Sustainable Expert Systems, ICSES 2024

作者： Ketineni, Sunilkumar Sheela, J. School of Computer Science and Engineering VIT-AP University Amaravati India

ISBN: (纸本)9798331540364

The management and extraction of pertinent information from sizable text collections depends critically on multi-document summarization. Using Transformer-based neural networks, this research describes a technical method for extractive multi-document summarization. To capture the important information across many publications, we suggest a unique architecture that makes use of self-attention processes and contextualized word representations. To improve the summary generation process, the model is trained using a combination of supervised learning and reinforcement learning methods. We run thorough tests on benchmark datasets and evaluate our method against current state-of-the-art techniques. The outcomes show how well our suggested model performs in producing informative and cogent summaries from a variety of documents. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Improving Cardiovascular Disease Detection: A Comparative approach with Boosting Algorithms and Imputed Features 1

Improving Cardiovascular Disease Detection: A Comparative Ap...

引用

1st International Conference for Women in Computing, InCoWoCo 2024

作者： Siddique Ibrahim, S.P. Madhu, Gifsy Kadali, Jayaprakash Salivendra, Nymisha Anitha, Badugu School of Computer Science and Engineering VIT-AP University Amaravati India

ISBN: (纸本)9798331518943

This paper is written as an approach to the prediction of cardiovascular diseases using a merged dataset from five sources, incorporating imputed 'Major Vessels' and 'Thallium' features to enhance predictive accuracy. Synthetic Minority over-sampling Technique (SMOTE) is applied to handle imbalance in classes, followed by eight feature selection methods (ANOVA, Chi-Squared, Random Forest, LASSO, RFE, etc.) to rank features based on selection frequency. Machine learning algorithms, including Logistic Regression, Random Forest, SVM, Gradient Boosting, and XGBoost, were trained and evaluated using various metrics. LightGBM achieved the best performance with a test accuracy of 94.61%, F1 Score of 0.9528, and ROC-AUC of 0.9747. Key features such as Thallium, ST slope, exercise induced angina, chest pain, major vessels and ST depression were consistently selected, highlighting their predictive importance. The study concludes that boosting algorithms are highly effective for heart disease classification, outperforming other models, particularly when more features are included. © 2024 IEEE.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

Stress Detection using Facial Emotion Recognition: Comparative Analysis of VGG-16, ResNet-50, and a Custom CNN Architecture 5

Stress Detection using Facial Emotion Recognition: Comparati...

引用

5th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2024

作者： Ibrahim, S.P. Siddique Chaladi, Praneetha Galam, Bhavya Sri Maridu, Bharathi Varsha Matta, Dhanalakshmi VIT-AP University School of Computer Science and Engineering Amaravati India

ISBN: (纸本)9798350389609

Stress is an important psychological factor that can significantly impact overall health and productivity;hence, timely detection and management of stress are essential to prevent adverse effects on mental and physical well-being. In this study, facial emotion recognition is employed to detect stress using the CK+ dataset, which is a widely used dataset for facial expression recognition, containing anger, disgust, contempt, fear, happiness, sadness, and surprise emotions. Initially, well-established architectures VGG-16 and ResNet-50 are tested on the CK+ dataset. Although these models delivered competitive accuracies, a custom hybrid model, based on Convolutional Neural Networks (CNN), achieved superior performance. The custom model used fewer parameters leading to faster training time while still maintaining a higher accuracy of 99%. The model outperformed already existing solutions in both accuracy and training time making it more effective for real-word stress detection through facial expressions. Notably, the custom model achieved comparatively higher accuracy than ResNet-50 and faster training time compared to VGG-16 model. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Cyber Sentinels: Illuminating Malicious Intent in Social Networks Using Dual-Powered CHAM 1

引用

2nd EAI International Conference on Pervasive knowledge and collective intelligence on Web and Social Media, PerSOM 2023

作者： Terumalasetti, Sailaja Reeja, S.R. School of Computer Science and Engineering VIT-AP University Amaravati India

ISBN: (数字)9783031660443

ISBN: (纸本)9783031660436

Online Social Networks (OSN), the security and reliability of these platforms are extremely vulnerable to malicious users. Online social networks’ volatile extension has amplified the pervasiveness of destructive practices comprising spamming, phishing, and disseminating false information. The administration of dynamic and altering antagonistic strategies has exposed complications in rule-based systems and anomaly detection techniques. Traditional rule-based approaches for detecting malicious behavior often fail to catch multifaceted and emergent threats. The emergent prominence of online social networks has made it essential to progress cutting-edge methods for spotting devious users and preserving network integrity. In this regard, the paper defines a distinctive method CHAM (CNN and Hierarchical Attention Mechanism) that enhances the detection of harmful traffic within these platforms by leveraging Convolutional Neural Networks (CNN) in conjunction with Hierarchical Attention Mechanism (HAM). Amalgam of both techniques enhances the benefit of the detection of malicious users in OSN precisely and efficiently. The model’s fundamental novelty is the adoption of the gated recurrent unit as the primary memory unit, coupled with layers for the attention mechanism, three degrees of maximum pooling, and layers for average pooling. These components work together to extract detailed flow characteristics, making it easier to identify subtle patterns suggestive of malicious behavior. A thorough data preparation phase is carried out before modeling to get precise data flow segments. The proposed framework takes the lead, promising improved detection effectiveness and a safer virtual world for all users. The methodology endeavors to elevate the precision and efficiency of malicious user detection. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Identification of Spam on Social Media by Semi-supervised Learning approach 1

引用

2nd EAI International Conference on Pervasive knowledge and collective intelligence on Web and Social Media, PerSOM 2023

作者： Kiran, Sudana Shashi Kiran, Sure Uday Amrutha, Elluru Yasaswi, E. Sai, Annam Kumar Sravya, Kusam Reeja, S.R. School of Computer Science and Engineering VIT-AP University Amaravati India

ISBN: (数字)9783031660443

ISBN: (纸本)9783031660436

This study examines how machine learning methods can be used to identify Twitter spammers. Due to spammers’ increased use of social media platforms, it is crucial to combat their fraudulent operations. This study uses extensive Twitter data, such as user profiles, tweet content, and network connections, to create algorithms that accurately differentiate between genuine users and spammers. By successfully identifying and filtering out spammers utilizing a range of machine learning algorithms, deep learning techniques, feature extraction methodologies, and even categorizing tweets based on their emotional tone, the ultimate goal is to improve platform trust and user experience. In order to efficiently identify spammers and fake users, the article provides a solution that makes use of a unique machine learning algorithm that meticulously analyses user personal information and account history. Because of its popularity, Twitter has drawn spammers and bogus users, which has caused confusion. The study seeks to address "Twitter Spam Drift" by applying a Semi-Supervised Learning approach (SSLA), which adjusts to changes in spam behavior and has demonstrated promising results on English datasets. To identify spam behaviors on Twitter, the project uses deep-learning classifier algorithms like LSTM, and GRU. Here, as can be seen, we use various methods to achieve accuracy ranging from 87% to 97%. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Improving security performance of Internet of Medical Things using hybrid metaheuristic model

引用

Multimedia Tools and applications 2025年第9期84卷 6403-6428页

作者： Kanneboina, Ashok Sundaram, Gopikrishnan Assistant Professor School of Engineering Malla Reddy University Telangana Hyderabad India School of Computer Science and Engineering VIT-AP University Andhra Pradesh Amaravathi India

The security of the Internet of Medical Things (IoMT) is very important because it connects many medical devices to improve care operations and patient monitoring in real time. Researchers have developed high-security frameworks with attribute-based access control, hashing, encryption, and privacy preservation. Current security models are either too complex or too unreconfigurable for real-time monitoring applications. Current models’ security performance depends on their internal configuration, which is not adaptable to new attacks. This research proposes an improved hybrid metaheuristic model to improve healthcare IoT security to overcome these restrictions. The model initially uses blockchain but can change its internal hashing and encryption standards. The proposed model continuously optimizes the blockchain-based IoMT deployment’s security and quality of service by combining Elephant Herding Optimisation (EHO) and Grey Wolf Optimisation. The proposed model uses dual fitness functions to improve security and service quality against various attacks. Multiple state-of-the-art models and DDoS, man-in-the-middle, masquerading, and Sybil attacks performed to evaluate the model. According to the comparison, the proposed model improved network consistency by 8.7%, throughput by 6.4%, packet delivery ratio by 8.2%, and attack detection and mitigation by 9.4%. Real-time healthcare use cases could use the suggested model after this performance. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Adaptive Neighborhood Adjustment Strategy Based On MOHHO and NSGA-III Algorithms for Feature Selection

引用

IAENG International Journal of applied Mathematics 2024年第5期54卷 917-935页

作者： Papasani, Anusha Durgam, Revathi Devarakonda, Nagaraju Computer Science and Engineering Department Koneru Lakshmaiah Education Foundation AndhraPradesh Vaddeswaram522502 India Computer Science and Engineering Department AVN Institute of Engineering and Technology Telangana Hyderabad501510 India Science and Engineering Department VIT-AP University AP Amaravati522237 India

The efficiency of multi-objective evolutionary algorithms (MOEAs) in tackling issues with multiple objectives is examined. However, it is noted that current MOEA-based feature selection techniques often converge towards the center of the Pareto front due to inadequate selection forces. The study proposes the utilization of a novel approach known as MOEA/D, which partitions complex multi-objective problems into smaller, more feasible single-objective sub-problems. Each sub-problem may then be addressed using an equal amount of computational resources. The predetermined size of the neighborhood used by MOEA/D may lead to a delay in the algorithm's merging and reduce the effectiveness of the failure. The paper proposes the Adaptive Neighbourhood Adjustment Strategy (ANAS) as a novel approach to improve the efficiency of multi-objective optimisation algorithms in order to tackle this issue. The ANAS algorithm allows for adaptive adjustment of the subproblem neighborhood size, hence enhancing the trade-off between merging and variety. In the following section of the study, a novel feature selection technique called MOGHHNS3/D-ANA is introduced. This technique utilizes ANAS to expand the potential solutions for a particular subproblem. The approach evaluates the chosen features using the Regulated Extreme Learning Machine (RELM) classifier on sixteen benchmark datasets. The experimental results demonstrate that MOGHHNS3/D-ANA outperforms four commonly employed multi-objective techniques in terms of accuracy, precision, recall, F1 score, coverage, hamming loss, ranking loss, and training time, error. The apBI approach in decomposition-based multi-objective optimization focuses on handling constraints by adjusting penalty parameters to guide the search towards feasible solutions. On the other hand, the ANA approach focuses on dynamically adjusting the neighborhood size or search direction based on the proximity of solutions in the detached space to adapt the search process.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Leveraging ResNet50 for Accurate Brain Tumor Diagnosis 2

Leveraging ResNet50 for Accurate Brain Tumor Diagnosis

引用

2nd International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2025

作者： Ranvitha, Gera Suhitha, Pureti Lakshmi Prasanna, Ghantasala Lakshmi Lasya, Avuthu Kumar, T. Bharath VIT-AP UNIVERSITY Andhra Pradesh Amaravati India School of Computer Science and Engineering VIT-AP UNIVERSITY Andhra Pradesh Amaravati India

ISBN: (纸本)9798331530389

Brain tumors are one of the foremost life-threatening diseases, producing serious complications that contribute to a heightened rate of morbidity and mortality. The emergence of deep learning strategies has vastly refined the process of identifying and categorizing brain tumors using MRI scans, providing crucial insights into the visualization of various structures within the human brain. Annually, countless individuals around the world receive a life-changing diagnosis of a brain tumor, and these cases often have low survival rates, underscoring the need for improved diagnostic approaches. This highlights the necessity for developing capable and systematic measures (improvement strategies) for diagnostics and treatment. While there has been progress in this regard, the search for a reliable detection system continues. In this paper, we focus on the ResNet50 model to enhance accuracy in detecting brain tumors. A type-curated ResNet50 is implemented for the brain tumor detection task, classifying MRI scans into two categories: 0 (Tumor) and 1 (No Tumor). Our findings reveal that the ResNet50 architecture successfully accomplished a classification accuracy of 97.3%, surpassing the VGG19 model, which attained an accuracy of 96%. This research seeks to contribute to the development of a robust and resilient model that enables healthcare professionals to make early and accurate decisions regarding brain tumors. © 2025 IEEE.

关键词： Brain Tumor deep learning ResNet50 VGG

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：