检索结果-内蒙古大学图书馆

2nd IEEE International Conference for Women in Innovation, Technology and Entrepreneurship, ICWITE 2024

作者： Sethi, Vikas Kumar, Rajneesh Mehla, Stuti Department of Computer Science and Engineering Mullana Ambala India Panipat Institute of Engineering and Technology Department of Computer Science and Engineering Panipat India

ISBN: (纸本)9798350383287

In 21st century, AI-based intelligent recommendation system uses rating predictions, which are frequently utilized and helps users swiftly filter down their options and make informed judgements from an abundance of material. Nevertheless, as recommender systems only contain a small number of explicit data and are subject to cold start and sparsity issues, the majority of existing recommendation models have poor accuracy. The majority of conventional algorithms used to make recommendations simply take into account the straight-forward connection among consumers and products, largely ignoring the interests, prospective employment factors, or other social aspects of the products being *** this study focus is on user preferences, as they are key to tailoring recommendations to individual needs. By combining trust values with user preferences, the system can recommend items that not only have a high trust value but also align with a user's individual tastes or needs. For this proposed model used feature matrix, Grey Wolf Optimizer (GWO) and Convolution Neural Network (CNN). GWO is an optimization algorithm used to find optimal weights or parameters for the recommendation system. Once the GWO has found optimal or near-optimal weights, they are integrated into the feature matrix. These weights can help prioritize or scale certain features to improve the recommendation quality. Then CNN will extract patterns and relationships from the feature matrix. Based on the processed feature matrix, the system predicts the top 'K' items to recommend to the user. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Hybrid golden jackal fusion based recommendation system for spatio-temporal transportation's optimal traffic congestion and road condition classification

引用

Multimedia Tools and Applications 2024年 1-34页

作者： Gawali, Tukaram K. Deore, Shailesh S. SSVPS’s B. S. Deore College of Engineering Maharashtra Dhule424002 India Department of Computer Engineering Government College of Engineering Maharashtra Jalgaon425001 India Department of Computer Engineering SSVPS’s B. S. Deore College of Engineering Maharashtra Dhule424002 India

Traffic congestion, influenced by varying traffic density levels, remains a critical challenge in transportation management, significantly impacting efficiency and safety. This research addresses these challenges by proposing an Enhanced Hybrid Golden Jackal (EGJ) fusion-based recommendation system for optimal traffic congestion and road condition categorization. In the first phase, road vehicle images are processed using Enhanced Geodesic Filtering (EGF) to classify traffic density as heterogeneous or homogeneous across heavy, medium and light flows using Enhanced Consolidated Convolutional Neural Network (ECNN). Simultaneously, text data from road safety datasets undergo preprocessing through crisp data conversion, splitting and normalization techniques. This data is then categorized into weather conditions, speed, highway conditions, rural/urban settings and light conditions using Adaptive Drop Block Enhanced Generative Adversarial Networks (ADGAN). In the third phase, the EGJ fusion method integrates outputs from ECNN and ADGAN classifiers to enhance classification accuracy and robustness. The proposed approach addresses challenges like accurately assessing traffic density variations and optimizing traffic flow in historical pattern scenarios. The simulation outcomes establish the efficiency of the EGJ fusion-based system, achieving significant performance metrics. Specifically, the system achieves 98% accuracy, 99.1% precision and 98.2% F1-Score in traffic density and road condition classification tasks. Additionally, error performance like mean absolute error of 0.043, root mean square error of 0.05 and mean absolute percentage error of 0.148 further validate the robustness and accuracy of the introduced approach. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Smoke Removal Method of Industrial Images Based on Dark Channel Prior Approach and Second-Generation Wavelets

引用

SN computer Science 2024年第7期5卷 1-10页

作者： Khmag, Asem Department of Computer Systems Engineering Faculty of Engineering University of Zawia Zawia Libya

In the field of digital image and computer vision, haze and smoke removal (dehazing) is one pf a popular scientific arena where it is being studied by an ample number of computer scientists. However, conventional joint dehazing and noise removal techniques use traditional hand-crafted approaches in order to achieve clean image by apply suitable restoration processes. In this study, a smoke removal technique, which designed according to dark channel, prior DCP is proposed. Theoretically, traditional DCP utilized the atmospheric light in the entire image as a global constant and as a result, in our study, the suggested DCP is further enhanced by using filter of second-generation wavelets (SGWs) and therefore, more powerful model is achieved. Consequently, the SGWs filtering is utilized due to its strength in custom designing for complex domains and irregular sampling in order to improve the solution of atmospheric light. The proposed algorithm is evaluated against several state of the art smoke removal methods. Extensive experimentation on multiple datasets demonstrates that our method exhibits better dehazing performance and generalizability than other image dehazing approaches in terms of peak signal-to-noise ratio PSNR, the structural similarity index SSIM and the average gradient results. Moreover, the execution efficiency of the proposed algorithm can reduce considerably the processing time in comparison with the state-of-the-art- smoke removal techniques. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Atmospheric light Dark channel prior Dehazing Patch size Second-generation wavelet Smoke removal

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Machine Learning, Ensemble Learning and Deep Learning Classifiers for Parkinson’s Disease Detection

引用

SN computer Science 2024年第1期5卷 1-14页

作者： Goyal, Palak Rani, Rinkle Department of Computer Science and Engineering Thapar Institute of Engineering and Technology Patiala India

A progressive neurodegenerative ailment called Parkinson's disease (PD) is marked by the death of dopamine-producing cells in the substantia nigra area of the brain. The exact etiology of PD remains elusive, but it is believed to involve the presence of Lewy bodies, abnormal protein aggregates, in affected brain regions, leading to the mobile symptoms of PD. Hence, as the management of PD continues to evolve, there is a growing demand for the establishment of a descriptive system that enables the early detection of PD. In this study, we conducted an extensive analysis using machine learning, ensemble learning, and deep learning models with different hyperparameters to develop accurate classification models for PD prediction. To enhance classifier performance and address overfitting, we employed principal component analysis (PCA) for feature selection along with various preprocessing techniques. The dataset used consisted of voice samples, comprising 188 PD patients and 64 normal individuals. Our results demonstrated that the Random Forest (RF) model with accuracy of 82.37% outperformed the other base classifiers Among the ensemble classifiers, the LGBM model exhibited the highest accuracy of 85.90% when compared to both base and ensemble classifiers. Notably, the deep learning model has 91.33% training accuracy and 85.02% testing accuracy, suggesting that deep learning models perform comparably equivalent on small datasets compared to machine learning classifiers. Overall, our findings underscore the effectiveness of machine learning, ensemble techniques and deep learning models in accurately predicting PD. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Deep learning Hyperparameters LGBM Machine learning Parkinson’s disease Random forest

来源：评论

学校读者我要写书评

暂无评论

Deep CNNs for glioma grading on conventional MRIs: Performance analysis, challenges, and future directions

引用

Mathematical Biosciences and engineering 2024年第4期21卷 5250-5282页

作者： Saluja, Sonam Trivedi, Munesh Chandra Saha, Ashim Department of Computer Science and Engineering National Institute of Technology Agartala Tripura 799046 India

The increasing global incidence of glioma tumors has raised significant healthcare concerns due to their high mortality rates. Traditionally, tumor diagnosis relies on visual analysis of medical imaging and invasive biopsies for precise grading. As an alternative, computer-assisted methods, particularly deep convolutional neural networks (DCNNs), have gained traction. This research paper explores the recent advancements in DCNNs for glioma grading using brain magnetic resonance images (MRIs) from 2015 to 2023. The study evaluated various DCNN architectures and their performance, revealing remarkable results with models such as hybrid and ensemble based DCNNs achieving accuracy levels of up to 98.91%. However, challenges persisted in the form of limited datasets, lack of external validation, and variations in grading formulations across diverse literature sources. Addressing these challenges through expanding datasets, conducting external validation, and standardizing grading formulations can enhance the performance and reliability of DCNNs in glioma grading, thereby advancing brain tumor classification and extending its applications to other neurological disorders. © 2024 the Author(s).

关键词： Magnetic resonance

来源：评论

学校读者我要写书评

暂无评论

A dataset for evaluating Bengali word sense disambiguation techniques

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第4期14卷 4057-4086页

作者： Das Dawn, Debapratim Khan, Abhinandan Shaikh, Soharab Hossain Pal, Rajat Kumar Department of Computer Science and Engineering University of Calcutta Calcutta India Product Development and Diversification ARP Engineering Calcutta India Department of Computer Science and Engineering BML Munjal University Kapriwas India

The computation of natural language enables a suitable transmission through the universe by retrieving the correct sense of each word. A word may be monosemous or polysemous. The use of polysemous words in an appropriate context plays a critical role in communication. Over the last 2 decades, a significant amount of research has been done for automatically solving the correct sense of a polysemous word in the context of word sense disambiguation. A word sense disambiguation algorithm identifies the proper sense of a polysemous word by analysing the contextual data. Nevertheless, there is a gap in the contemporary literature regarding the availability of datasets in Asian languages, especially Bengali. Therefore, in this work, we have presented a dataset comprising hundred Bengali polysemous words. Each word in this dataset consists of three or four disjoint senses, and each sense comprises ten paragraphs. Each paragraph describes the sense of a particular polysemous word. We have performed statistical analysis on the basis of seven relevant and important characteristics. A general framework has also been presented for training and testing with possible guidelines for performance analysis. A baseline strategy has been introduced based on four feature sets. Finally, a set of experiments have been performed to analyse the system performance. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Dual Watermarking for Image Copyright Protection and Authentication

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第12期5卷 6134-6145页

作者： Padhi, Sudev Kumar Tiwari, Archana Ali, Sk. Subidh Indian Institute of Technology Bhilai Department of Computer Science and Engineering Durg491002 India

Advancements in digital technologies make it easy to modify the content of digital images. Hence, ensuring digital images' integrity and authenticity is necessary to protect them against various attacks that manipulate them. We present a deep learning (DL) based dual invisible watermarking technique for performing source authentication, content authentication, and protecting digital content copyright of images sent over the internet. Beyond securing images, the proposed technique demonstrates robustness to content-preserving image manipulation attacks. It is also impossible to imitate or overwrite watermarks because the cryptographic hash of the image and the dominant features of the image in the form of perceptual hash are used as watermarks. We highlighted the need for source authentication to safeguard image integrity and authenticity, along with identifying similar content for copyright protection. After exhaustive testing, our technique obtained a high peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM), which implies there is a minute change in the original image after embedding our watermarks. Our trained model achieves high watermark extraction accuracy and satisfies two different objectives of verification and authentication on the same watermarked image. © 2020 IEEE.

关键词： Watermarking Copyright Protection Authentication Robustness Decoding Deep Learning Artificial Intelligence Accuracy PSNR Feature Extraction Copyright Protection Cryptographic Hash Deep Learning Image Authentication Perceptual Hash Watermarking

来源：评论

学校读者我要写书评

暂无评论

CVApool: using null-space of CNN weights for the tooth disease classification

引用

Neural Computing and Applications 2024年第26期36卷 16567-16579页

作者： Can, Zuhal Isik, Sahin Anagun, Yildiray Department of Computer Engineering Eskisehir Osmangazi University Meselik Campus Eskisehir26489 Turkey

In light of current developments in dental care, dental professionals have increasingly used deep learning methods to get precise diagnoses of oral problems. Using intraoral X-rays in dental radiography is imperative in many dental interventions. Integrating deep learning techniques with a unique collection of intraoral X-ray images has been undertaken to enhance the accuracy of dental disease detection. In this study, we propose an alternative pooling layer, namely the Common Vector Approach Pooling technique, to address the constraints associated with average pooling in deep learning methods. The experiments are conducted on a large dataset, involving twenty different dental conditions, divided into seven categories. Our proposed approach achieved a high accuracy rate of 86.4% in identifying dental problems across the seven oral categories. © The Author(s) 2024.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

CAAM- CNN With Autoencoder Attention Mechanism for Recommendation System to Improve Trust in E-commerce Industry 2

CAAM- CNN With Autoencoder Attention Mechanism for Recommend...

引用

2nd IEEE International Conference for Women in Innovation, Technology and Entrepreneurship, ICWITE 2024

作者： Sethi, Vikas Kumar, Rajneesh Mehla, Stuti Department of Computer Science and Engineering Ambala Mullana India Panipat Institute of Engineering and Technology Department of Computer Science and Engineering Panipat India

ISBN: (纸本)9798350383287

COVID-19 crisis expedited the spread of e-commerce, moving beyond luxury goods and services and towards new businesses, product categories and basic necessities that are important to many people. Customers can have access to a wide range of products from the comfort and security of their homes but still have trusting issues while purchasing online. Online feedback either comments or ratings given by customers, has the power to affect how consumers make purchasing decisions. This meteoric expansion of digitized information and number of users over internet brought the use of recommender systems as a significant solution by offering more proactive and individualized information services and enable users to quickly narrow their choices and make wise ***, the majority of existing recommendation models have poor accuracy issues as a result of data scarcity and cold start issues because recommender systems only have a limited amount of explicit data and this also lead to trust issue where users are fear of being taken advantage due to misleading information on the e-commerce websites. To solve this problem and provide quality recommendations, a new trust-based approach CNN with Autoencoder Attention Mechanism (CAAM) is proposed. CAAM uses static and dynamic trust combined with the user-item preferences and pass through the auto-encoder to reduce noise and make encoded features visible. After that convolution is used to combine encoded features with user and item features in latent space to provide nonlinear mapping and increase the reliability of features and to improve the recommendations which ultimately lead to improve the trust of people over *** Epinions dataset is used to evaluate the concept in a number of measures. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Blockchain based intelligent disbursement in National Scholarship Portal

IET Blockchain

引用

IET Blockchain 2024年第4期4卷 407-422页

作者： Samu, Lifna Challissery Khithani, Neelkanth Alve, Kushl Gambhire, Vedang Hande, Atharva Choubey, Shivam Department of Computer Engineering Vivekanand Education Society's Institute of Technology Mumbai India

The National Scholarship Portal in India serves as a one-stop solution for students seeking financial aid for their studies across the country. However, in this digital era, the national-level portal faces challenges such as limited provision for only government scholarships, non-automated systems, complex application processes, reliance on physical verifications, and delays in scholarship disbursement. This research proposes a blockchain-based scholarship module to address these challenges and automate the entire scholarship process. The paper emphasizes upon the transformative impact by the usage of Hyperledger fabric network, which provides a fool-proof system that streamlines the entire application process and fund disbursement. The proposed integration also ensures robust application verification, accountability of stakeholders, transparent scholarship selection criteria, automated and thorough tracking of fund disbursement, immutable transaction history, secure authorization;and stringent compliance measures. Thus, the implementation of the proposed system aims to alleviate the financial insecurities faced by students during their studies, simplify their search for scholarship opportunities, and enable them to focus more on their academic pursuits. © 2024 The Author(s). IET Blockchain published by John Wiley & Sons Ltd on behalf of The Institution of engineering and Technology.

关键词： Blockchain

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：