检索结果-内蒙古大学图书馆

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

作者： Sai, Gurram Kumar Harshith, Koganti Sri Sai Varma, Kucharlapati Akash Yashwanth Kiran, G. Swaminathan, J. Amrita Vishwa Vidyapeetham Department of Computer Science and Engineering Amritapuri India

ISBN: (纸本)9798350370249

In this research, we delve into a transformative exploration of the realm of code generation, unveiling three fundamental modules that constitute the backbone of our study: the Text-to-Code Generation Model, Code Autocompletion Model, and the formidable Neural Program Synthesis Model. At the forefront of our investigation, we underscore the paramount importance of Neural Program Synthesis - a dynamic force adept at crafting custom programs from input-output examples. Our exploration delves into the intricate landscape of natural language-to-code generation, offering profound insights and invaluable contributions to the field. The ability to automatically generate code from natural language instructions holds immense significance in streamlining software development processes, reducing manual effort, and enhancing productivity. Furthermore, this research addresses a pressing need within the software engineering community for efficient methods of code generation, particularly in domains where human-program interaction is prevalent. Through rigorous experimentation and meticulous analysis, we shed light on methodologies and promising avenues for future research in this burgeoning domain, paving the way for advancements in automated software development and programming assistance tools. © 2024 IEEE.

关键词： Input output programs

来源：评论

学校读者我要写书评

暂无评论

FLARE: Federated Learning And Resilient Encryption for Firewalls 7

FLARE: Federated Learning And Resilient Encryption for Firew...

引用

7th IEEE Pune Section International Conference, PuneCon 2024

作者： Senthilkumar, Lohith Rengarajan, Aaditya Psg College of Technology Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798331527822

Traditional firewalls rely on static rule-based mechanisms, where rules are manually defined and often written in specialized languages. While effective to a degree, these rules are inherently limited and can be easily bypassed by new and evolving types of malware, leading to significant security vulnerabilities. To address these challenges, we propose FLARE: Federated Learning And Resilient Encryption, a novel machine learning-based firewall solution. FLARE dynamically analyzes past network connections to predict and determine the appropriate actions for incoming traffic, thereby adapting to emerging threats in *** the sensitive nature of firewall data, which often contains confidential information, the training of machine learning models poses significant privacy risks. FLARE mitigates these risks by incorporating federated learning, allowing the model to learn from decentralized data sources without requiring raw data to be shared. To further enhance privacy, we introduce an encryption layer that ensures the central model learns from encrypted weights, preventing exposure of sensitive information known to the local model. This combined approach not only improves the resilience of firewalls but also safeguards the confidentiality of the data used in training, offering a robust solution for modern cybersecurity challenges. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Design of high output DC/DC boost circuit without boost transformer 8

Design of high output DC/DC boost circuit without boost tran...

引用

8th International Conference on Communication and Information Systems, ICCIS 2024

作者： Hong, Qin Zhou, Liang Jia, Zhongning Liu, Chun Liu, Jianbo Software Engineering Institute of Guangzhou Department of Computer Science Guangzhou China

ISBN: (纸本)9798331531171

The development of IOT smart devices has necessitated increasingly stringent requirements for DC power sources in terms of output voltage, size, and cost. For instance, Avalanche Photodiodes (APDs), utilized as photodetectors, require a significantly high reverse bias voltage during the detection of feeble optical signals to achieve optimal gain and high signal-to-noise ratio, thereby ensuring exceptional sensitivity. To address the diverse demands of various smart devices for DC power, a high-output DC/DC boost circuit without a step-up transformer has been innovatively designed. This circuit is comprised of four fundamental components: an oscillator circuit, a comparator circuit, a voltage multiplier circuit, and a constant-current and voltage-stabilizing circuit. The paper conducts a comprehensive theoretical analysis of the circuit's topology, parameter selection, and operational process, which is further validated through Multisim simulation. Additionally, a PCB circuit was designed, fabricated, and underwent rigorous practical testing. The test results conclusively demonstrate that the designed boost circuit proficiently provides APDs with the appropriate bias voltage, eliminating the need for boost transformer. It boasts high output voltage, a simplistic structure, small size, and low cost, enabling it to seamlessly share the same input power source with the APD detection circuit and be integrated into the same PCB layout. © 2024 IEEE.

关键词： avalanche photodiode boost circuit direct current stabilized power supply oscillator voltage multiplier circuit

来源：评论

学校读者我要写书评

暂无评论

Evaluating Educational Commentaries: Sentiment Analysis for Student Testimonials Using Machine Learning 4

Evaluating Educational Commentaries: Sentiment Analysis for ...

引用

4th International Conference on Technological Advancements in Computational sciences, ICTACS 2024

作者： Kasana, Tushar Kaur, Baljinder Rakhra, Manik Lovely Professional University Department of Computer Science and Engineering Phagwara India

ISBN: (纸本)9798350387490

This is because in education, evaluation of student data especially in form of feedback cannot be complete without sentiment analysis. The various methods adopted in analysis of sentiments in educational context are discussed in this paper though the use of deep learning and machine learning approaches. Prominent findings show how effectively the mentioned strategies may be implemented to enhance instructional practices and shape institutional decisions. An example of concepts described is such problems as biased datasets or conflict of cultural contexts. The study reveals how self-service feedback tools such as sentiment analysis tools can revolutionize feedback receipt as well as instruction, paving way for individualized learning environment and enhanced performance outcomes. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Skin Lesion Segmentation Using Deep Learning 2nd

Skin Lesion Segmentation Using Deep Learning

引用

2nd International Conference on Cyber Intelligence and Information Retrieval, CIIR 2023

作者： Sarkar, Kaustav Haralalka, Varun Shaw, Vishal Singh, Rishav Raj Jaiswal, Rayirth Pramanick, Tridip Department of Computer Application and Science Institute of Engineering and Management Kolkata India

ISBN: (纸本)9789819735938

Dermoscopic image analysis has gained significant importance for dermatological applications due to its potential in eliminating observer bias. Segmenting lesion areas automatically in these images is crucial for expedited disease diagnosis. In this study, an innovative deep learning-based approach for precise lesion delineation has been proposed. Leveraging a deep convolutional neural network, where input images are included after appropriate preprocessing. Pedro Hispano Hospital (PH2) 2013 dataset has been employed, comprising dermoscopic images and lesion binary masks as ground truth for evaluation. The approach uses the SegNet architecture, which achieves remarkable accuracy of 99.14%, surpassing benchmark standards (DC 96.94, JI 99.11).This research offers a promising solution for automated lesion segmentation in dermoscopic images, advancing the field of medical image analysis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Image analysis

来源：评论

学校读者我要写书评

暂无评论

UAV and UGV Assisted Path Planning for Sensor Data Collection in Precision Agriculture 11

UAV and UGV Assisted Path Planning for Sensor Data Collectio...

引用

11th International Symposium on Electronic Systems Devices and Computing, ESDC 2023

作者： Singh, Himanshu Biken Singh, Moirangthem Pratik, Harsh Pratap, Ajay Department of Computer Science and Engineering Varanasi India

ISBN: (纸本)9781665455725

In recent years, automated (Intelligent) decision support systems have become prevalent in various smart city applications such as healthcare, transportation, energy management, and environmental monitoring. Internet of Things (IoT) and Unmanned Aerial Vehicle (UAV)-based smart sensing and actuation devices in agricultural events can change the sector from static and manual to dynamic and intelligent, resulting in increased production with minimal human efforts. In addition, soil measurements that are time-consuming can be collected using Unmanned Ground Vehicles (UGVs). However, to collect data efficiently from wireless sensors in agricultural fields, UAV and UGV need to follow an optimal path. Thus, in this paper, we formulate utility maximization problem using UAV and UGV by simultaneously minimizing energy consumption and maximizing data collection. To solve the formulated problem, we propose a modified Greedy Randomised Adaptive Search Procedure (GRASP) algorithm to predict an efficient path for UAV and UGV to collect data from the agricultural field. Moreover, the efficacy of the proposed algorithm is showcased theoretically and experimentally on real-world data and compared with other state-of-the-art methods. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Automatic Number Plate Recognition System using Machine Learning Techniques

Automatic Number Plate Recognition System using Machine Lear...

引用

2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024

作者： Dayal, Avnish Porwal, Aryan Gupta, Avinash Srivastava, Aditya Kumar Mahendru, Mansi ABES Engineering College Department of Computer Science UP Ghaziabad India

ISBN: (纸本)9798350378092

The system employs video processing technology to detect the image of a vehicle's number plate from a video stream. Subsequently, various algorithms are applied to extract the number from the detected plate. This system can be used at college entrances, cantonment area and other places with strict access restrictions. The system detects the number plate whenever a vehicle passes by, converting it into both images and text. Subsequently, this information is stored in the database for future reference or analysis. This research paper introduces ANPR system which is designed to optimize accuracy, efficiency, and ease of implementation which includes combination of Python programming language, YOLOv8 (You Only Look One-level), and EasyOCR, the proposed system addresses the complexities associated with detection of license plate and recognition in diverse real-world situations. Core of the system lies in the adoption of YOLOv8, an object detection model renowned for its real-time capabilities and high accuracy. EasyOCR, a user-friendly Optical Character Recognition (OCR) library, is seamlessly integrated into the ANPR system. This addition enhances the system's ability to accurately extract alphanumeric information from detected license plates, bridging the gap between detection and recognition phases. Due to this the system ensures efficient noise reduction, image segmentation, and overall improvement in the quality of input images, contributing to the system's reliability. The paper presents an in-depth evaluation of the ANPR system's performance through datasets. Additionally, the research explores potential applications of the ANPR system in real-world scenarios, emphasizing its applicability in traffic management, law enforcement, and security systems. © 2024 IEEE.

关键词： Noise abatement

来源：评论

学校读者我要写书评

暂无评论

Multimodal Hate Speech Detection using Fine-Tuned Llama 2 Model

Multimodal Hate Speech Detection using Fine-Tuned Llama 2 Mo...

引用

2024 International Conference on Intelligent Algorithms for Computational Intelligence Systems, IACIS 2024

作者： Sasidaran, Keerthana Geetha, J. Ramaiah Institute of Technology Department of Computer Science and Engineering Bangalore India

ISBN: (纸本)9798350360660

Hate speech detection is crucial for secure online environments, especially with recent policy changes on social media. This study enhances the Llama 2 model by fine-tuning it with a comprehensive Twitter dataset to improve hate speech detection accuracy. The system's frontend allows users to upload or record audio, upload videos, or provide YouTube URLs, with processing managed by the Google Speech Recognition API and the Whisper library. The backend integrates advanced speech recognition and video processing technologies. The refined Llama 2 model achieved an F1 score of 0.94, with accuracy, precision, and recall metrics surpassing other state-of-the-art models. These results demonstrate the model's high effectiveness and reliability in hate speech detection. This research highlights the potential of transformer-based models in enhancing online safety and inclusivity. Future work will focus on expanding the dataset, improving model generalization, supporting additional languages, and enhancing system scalability to better handle evolving hate speech forms. © 2024 IEEE.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

STPM: Spatial-Temporal Token Pruning and Merging for Complex Activity Recognition

引用

IEEE Transactions on Circuits and Systems for Video Technology 2024年第6期35卷 5407-5418页

作者： Su, Yumeng Zhang, Jiachao Yan, Rui Li, Pengpeng Xie, Guosen Shu, Xiangbo Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China Nanjing Institute of Technology Artificial Intelligence Industrial Technology Research Institute Nanjing211167 China Nanjing University Department of Computer Science and Technology Nanjing210023 China

Lightweight video representation techniques have advanced significantly for simple activity recognition, but they still encounter several issues when applied to complex activity recognition: (i) The presence of numerous individuals and varying spatial positions makes it difficult for traditional token pruning methods to maintain accuracy. (ii) Simply discarding entire frames may result in the loss of crucial clues. (iii) To maintain parallel computing, applying the same pruning rate to every frame leads to significant redundancy in frames with low information content. To this end, we propose a lightweight and novel Spatial-Temporal Token Pruning and Merging (STPM) framework, specifically designed for complex action videos where human actors occupy a small spatial resolution within video frames. Our framework considers two critical factors: semantic importance and spatial-temporal redundancy, to further reduce overhead. For semantic importance, STPM captures class-specific attention scores by learning multiple class tokens within the transformer to guide token pruning. For spatial-temporal redundancy, STPM employs an anchor graph and temporal attention to perform spatial and temporal token merging, preserving appearance and temporal cues while eliminating semantic duplication and redundancy. We conduct extensive experiments on JRDB-PAR primarily using recently introduced video transformer backbones, e.g., MViT and ViT. Our framework achieves similar results while requiring 40% less computation. © 1991-2012 IEEE.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

FLoomChecker: Repelling Free-riders in Federated Learning via Training Integrity Verification 30

FLoomChecker: Repelling Free-riders in Federated Learning vi...

引用

30th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2024

作者： Liang, Guanghao Chang, Shan Dai, Minghui Zhu, Hongzi Donghua University School of Computer Science and Technology Shanghai China Shanghai Jiaotong University Department of Computer Science and Engineering Shanghai China

ISBN: (纸本)9798331515966

Federated learning is a mechanism that allows participating clients to train locally with their own data in order to receive rewards, thus avoiding the transfer of data to a central server and protecting users' privacy. However, some 'lazy' clients may adopt the strategy of fabricating false model local updates in an attempt to 'free-riding' without actually contributing real data or consuming local computational resources. To address this issue, we propose FLoomChecker, an integrity detection scheme for federated learning training models. The scheme combines the techniques of trusted execution environments and Bloom filters to efficiently identify clients that do not train honestly by committing and proving. We conducted experimental evaluations of FLoomChecker, examining three main aspects: query time, build time, and memory footprint in trusted execution environment (TEE). The experimental results demonstrate the effectiveness of our scheme, and its performance improves as the number of local training rounds increases. © 2024 IEEE.

关键词： Federated learning

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：