检索结果-内蒙古大学图书馆

5th International Conference on Smart Systems and Inventive Technology, ICSSIT 2023

作者： Raveena, S. Surendran, R. Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9781665474672

up of coffee is a part of the daily human routine life. Any small changes in the coffee cherries yield that leads a huge change in coffee quality, smell, and taste. Types and stages of coffee cherries Identification and categorization of varieties and stages necessitate the presence of a farmer or crop pathologist. Manual illness diagnosis might result in mistaken identity and require a lengthy process. This work presents a comparison between the Deep Convolutional Neural Network-based approach and ResNet50 for classifying coffee type and maturity using transfer learning. Coffee cherries of different stages and varieties may be automatically sorted and recognized. Many ways to solve this problem have been proposed with transfer learning emerging as the dominant option due to its outstanding performance. This research work has implemented the Deep CNN Transfer Learning to detect distinct phases in the coffee cherries datasets. The suggested model employs VGG 16 for classifying coffee type and maturity got 96.38% accuracy and using ResNet50 for classifying coffee type and maturity with 99.01% accuracy. Experiments demonstrate that the suggested strategy is practical and that it would be used to detect efficiently and outperform other methodologies. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Harnessing LLMs for Medical Insights:NER Extraction from Summarized Medical Text 15

Harnessing LLMs for Medical Insights:NER Extraction from Sum...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Sasikala, D. Sudarshan, R. Sivasathya, S. Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Engineering Chennai India Pondicherry University School of Engineering & Technology Department of Computer Science Pondicherry India

ISBN: (纸本)9798350370249

The rapid growth of medical text presents a significant challenge for healthcare professionals to stay updated with current research and clinical guidelines. Effective text summarization is essential to facilitate access to knowledge, improve clinical decision-making, and ultimately enhance patient outcomes. This study aims to fine-tune and assess a medical text summarization system utilizing large language models (LLMs). The objective is to produce concise, coherent summaries of medical literature, prioritizing clinical relevance. Additionally, automated named entity recognition (NER) is employed to extract vital entities from the summaries generated by GPT3, T5, PEGASUS LLMs. These LLMs are trained on benchmark PubMed dataset which are evaluated under standard metrics such as ROUGE (1, 2, L) and BERTScore metrics. The model pipeline demonstrates summarization and NER performance over precision of 6 5% on medical text data. The resulting entity-annotated summaries provide condensed representations, enabling efficient information extraction and knowledge discovery. This automated approach accelerates biomedical research by large-scale medical text analysis. © 2024 IEEE.

关键词： GPT3 large language model medical text summarization Named Entity Recognition PEGASUS T5

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Image Captioning Techniques 4

A Comparative Analysis of Image Captioning Techniques

引用

4th International Conference for Emerging Technology, INCET 2023

作者： Sunil, Diya Theresa Safar, Seema Das, Abhijith Amijith, M.M. Joshy, Devika M. Rajagiri School of Engineering and Technology Department of Computer Science and Engineering India

ISBN: (纸本)9798350335750

Image captioning is the task of generating a textual description that accurately represents the content of an image. This task involves combining computer vision techniques, such as object recognition and scene understanding, with natural language processing to produce a human-like description of an image. Over time, various models have been introduced to perform image captioning, all aiming to accurately describe the content of an image. These models have practical applications such as improving the accessibility of multimedia content, assisting individuals with visual impairments, medical image captioning, and enhancing image search and retrieval. This paper explores some of the models and studies their efficiency using different evaluation metrics. © 2023 IEEE.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Deep Neural Network-based Crime Scene Detection with Frames 8

Deep Neural Network-based Crime Scene Detection with Frames

引用

8th IEEE International Conference on Science, Technology, engineering and Mathematics, ICONSTEM 2023

作者： Nandhini, J.T. Thinakaran, K. Department of Computer Science and Engineering Saveetha School of Engineering Chennai India

ISBN: (纸本)9798350347791

The field of computer vision stands to benefit significantly from automated crime scene detection. In this work, we demonstrate the application of DNN (Deep Neural Network) to identify a knife, blood, and gun in a picture and forecast whether a crime has occurred. Improving the accuracy of the system's detection capabilities was a top priority for us since we wanted to make the system as helpful as possible. To get a detection result, DNN's Non-linearity ReLu (Rectified Linear Unit), a Convolutional Neural Network Layer, a Fully Connected Layer, and a Dropout Function are employed. Additionally, this research makes use of a Dropout Function. To get the outcomes we want from Neural Networks, we use the open-source platform TensorFlow. Our system has a test accuracy of about 92.1% for the datasets that are currently accessible, which puts it in a very competitive position compared to other systems specifically created for this task. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Breaking classification Algorithms: A Black-Box Adversarial Attack for Data Poisoning

Breaking classification Algorithms: A Black-Box Adversarial ...

引用

2023 International Conference on Innovations in engineering and Technology, ICIET 2023

作者： Doss P, Lourdu Mahimai Gunasekaran, M. Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350309980

Poisoning attacks interfere with machine learning classifiers' performance by introducing malicious data into the learning dataset. In this paper, we propose a black-box adversarial attack for data poisoning in machine learning, where the attacker has no knowledge of the classifier's internal parameters or training data. Our attack utilizes a novel poison threat algorithm that leverages the gradient descent method to generate adversarial examples that deceive the classifier. We evaluate our attack on two popular classification algorithms, decision trees and support vector machines, using the random circles dataset. According to our results, the proposed attack has the ability to compromise the performance of both classifiers, highlighting the need for robust countermeasures against poisoning attacks. © 2023 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Predictive Data Analysis to Support Decision-Making Based on Long-Term Impacts of Disasters 1

引用

5th International Conference on computer and Communication Technologies, IC3T 2023

作者： Mishra, Shailendra Kumar Rahamatkar, Surendra Amity School of Engineering and Technology Department of Computer Science and Engineering Amity University Chhattisgarh Raipur India

ISBN: (数字)9789819997077

ISBN: (纸本)9789819997060

Research into disaster relief is a field with a lot of potential, and predictions made about its future are based on historical data. It is a difficult task to analyze the long-term consequences of a disaster. Governments, NGOs, and other organizations struggle to make sense of disaster data. It is hard to assess the impact of different factors, such as education, employment, agriculture, and the social sector, on people living in affected areas. This research paper represents a framework to evaluate the long-term effects of disasters, by means of machine learning algorithms. The algorithms mentioned in this paper are Random Forest, Decision Tree, Gradient-boosted Decision Tree (GBDT), and XGBoost. It seeks to analyze historical disaster data, identify impacts, and predict the effects of similar disasters in the future. The framework also suggests preemptive measures and the impact on those involved. Finally, the proposed framework will be able to make long-term analyses and recommendations, using data from past events. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Disasters

来源：评论

学校读者我要写书评

暂无评论

Enhancing Fault Tolerance in Wireless Mesh Networks through Adaptive and Resilient Routing Protocols

Enhancing Fault Tolerance in Wireless Mesh Networks through ...

引用

2023 International Conference on Data Science, Agents and Artificial Intelligence, ICDSAAI 2023

作者： Caleb, S. Thangaraj, S.John Justin Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350348910

From smart cities to disaster response systems, Wireless Mesh Networks (WMNs) are an essential component of many different kinds of infrastructure. Their great sensitivity to different types of errors, however, makes it very difficult to maintain trustworthy communication. The goal of this research is to improve WMN fault tolerance by developing new resilient and adaptive routing algorithms. The main goal is to detect faults in real-time using dynamic algorithms that can detect and react to disturbances including link failures, node unavailability, and changes in the environment. On top of that, the research suggests adaptive routing algorithms that may respond to changing network conditions, improve communication pathways, and avoid broken parts. In the meanwhile, we look at resilient topology control methods, which stress the need to keep a strong network architecture with features like redundancy, load balancing, and quick adaptability to topological *** research takes a novel tack by combining machine learning models for fault prediction and proactive repair. These models add to a predictive fault-tolerance paradigm by using network patterns and historical data, which further strengthens the network's resilience. Included in the thorough evaluation of the protocols created in this paper are simulations and real experiments. These evaluations cover important performance metrics like latency, throughput, and overall network reliability. The study reveals the suggested protocols' strengths and weaknesses by a thorough comparison with current fault-tolerant routing methods. By laying the groundwork for resilient and adaptive fault-tolerant communication in complex and ever-changing contexts, the results ultimately help wireless mesh networks progress. © 2023 IEEE.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

Enhancing the Robustness of Deep Neural Networks using Deep Neural Rejection 2

Enhancing the Robustness of Deep Neural Networks using Deep ...

引用

2nd IEEE International Conference on Advances in Computing, Communication and Applied informatics, ACCAI 2023

作者： Doss P, Lourdu Mahimai Gunasekaran, M. Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350315905

Adversarial examples are inputs that have been intentionally generated in order to deceive deep neural networks (DNNs) into generating inaccurate predictions. These instances endanger the security and safety of DNNs in real-world applications. To solve this problem, we present a new defense against adversarial instances based on Deep Neural Rejection (DNR). The DNR approach involves training a secondary model, referred to as the rejector model, to identify and reject inputs that are unlikely to produce correct predictions. The rejector model can be trained using adversarial examples as well as benign samples to learn the difference between the two types of inputs. If an input is rejected by the rejector model, it is assumed to be adversarial, and the primary model will not make a prediction on it. The experimental findings show that the DNR technique improves DNN resilience against hostile cases while retaining excellent accuracy on benign samples. Furthermore, by lowering the amount of samples that must be processed, the DNR technique can minimize the computational cost of DNNs. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Analysis of Permeability Characteristic of Planar Antenna with Spiral Resonator (SR) Shape

Analysis of Permeability Characteristic of Planar Antenna wi...

引用

2024 Photonics and Electromagnetics Research Symposium, PIERS 2024

作者： Tan, Yamato Wismiana, Evyta Yunus, Mochamad Prasetyo, Agus Dwi Haidi, Junas Munir, Achmad University of Pakuan Faculty of Engineering Department of Electrical Engineering Indonesia Telkom University School of Electrical Engineering Indonesia Institut Teknologi Bandung Radio Telecommunication and Microwave Laboratory School of Electrical Engineering and Informatics Indonesia University of Bengkulu Faculty of Engineering Department of Electrical Engineering Indonesia

ISBN: (纸本)9798350375909

Structural analysis of antenna configuration is becoming prominent for drawing up behavioral characteristics of the antenna. This paper deals with the permeability characteristic analysis of planar antenna configured by a spiral resonator (SR) shape. The analysis is performed by modeling the SR shape planar antenna using its equivalent circuit to formulate the permeability (μ) of antenna. To calculate the parameter of parts in the equivalent circuit, including resistance, capacitance, and inductance, the formula of Clausius-Mossotti suitable for SR structure are implemented, thereby obtaining the effective impedance of antenna. Based on these formulations, behavioral characteristics of the antenna are observed, and its permeability characteristic as the frequency dependency is then visualized. It shows that the effective permeability has the real value of -2.5 at the lowest frequency of 0.1GHz, which is relatively unchanged to the number of spiral turns. Meanwhile, the imaginary part shows the frequency trend shifts to the lower frequency as the number of spiral turns increases. © 2024 IEEE.

关键词： Equivalent circuits

来源：评论

学校读者我要写书评

暂无评论

A novel transfer deep learning method for detection and classification of plant leaf disease

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第9期14卷 12407-12424页

作者： Kaur, Prabhjot Harnal, Shilpi Gautam, Vinay Singh, Mukund Pratap Singh, Santar Pal Chitkara University Institute of Engineering and Technology Chitkara University Punjab India School of Computer Science Engineering and Technology Bennett University Greater Noida India Department of Computer Science and Engineering Rashtrakavi Ramdhari Singh Dinkar College of Engineering Begusarai India

The major cause of plant mortality and devastation, particularly among trees is plant diseases. This problem, however, may be handled and treated effectively through early detection. Crop/plant diseases must be identified and prevented as soon as possible to improve yield. Plant disease classification and identification by plant protection professionals through eye observation is a time-consuming and error-prone process. Previously, many researchers used hand-crafted techniques for the detection and classification of diseases. Also, due to a lack of knowledge experts and improper laboratory facilities, classification at an early stage is not possible. In terms of precision agriculture applications, developing effective, rapid, and highly successful computer-aided disease detection systems have become a necessity. A deep learning-based convolutional neural network (CNN) technique is used successfully in numerous computer vision tasks such as picture classification, object detection, segmentation, and image analysis for finding the best results. The existing models had been worked on a mixed crop approach and single class dataset with average accuracy. In this paper, the Modified InceptionResNet-V2 (MIR-V2) which is a form of the CNN model is used in conjunction with a transfer learning approach to recognize illnesses in images of tomato leaves. The proposed model is trained on a public dataset and self-collected dataset that includes seven different tomato leaf disease classifications with one healthy leaf. For the evaluation of model performance, different parameters such as dropout, learning rate, batch size, count of epochs, and accuracy are used. The disease classification accuracy rate for the applied models is 98.92% and the F1 score is 97.94%. The experimental results support the suggested approach’s validity and find it effective in detecting tomato leaf disease. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer

关键词： Classification (of information)

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：