检索结果-内蒙古大学图书馆

4th International Conference on Advanced Research in Computing, ICARC 2024

作者： Muthaliff, Faizan Sumanathilaka, T.G.D.K. School of Computing Informatics Institute of Technology Colombo06 Sri Lanka Swansea University Department of Computer Science Wales Swansea United Kingdom

ISBN: (纸本)9798350384864

In this day and age where convenience is king a new competitor in the form of chat-bots has appeared which requires users to only have prior knowledge on the language the chat-bot converses in. When it comes to planning outings there are a multitude of criteria to be wary of to determine the ideal solution. This review paper will address the common issues with extracting meaningful information from the user utilizing modern NLP techniques, the required criteria in addition to the user input such as, real time weather information and communicable diseases pertaining to the outing location, to determine an ideal suggestion and finally the method in which said ideal suggestion is determined after prioritizing criteria. The conclusion underscores that while NLP applications might not entirely replace existing menu-based systems for outing planning, they excel in certain scenarios. Notably, NLP's innate capacity to comprehend user preferences and context allows it to offer tailored outing suggestions. Furthermore, the application's unique feature of dynamically adapting to user requests without necessitating a return to the home page adds a distinctive advantage to its usability. © 2024 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Mitigating Sycophancy in Large Language Models via Direct Preference Optimization

Mitigating Sycophancy in Large Language Models via Direct Pr...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Khan, Azal Ahmad Alam, Sayan Wang, Xinran Khan, Ahmad Faraz Neog, Debanga Raj Anwar, Ali University of Minnesota-Twin Cities Department of Computer Science and Engineering United States Indian Institute of Technology Guwahati India Virginia Tech Department of Computer Science United States

ISBN: (纸本)9798350362480

Large language models (LLMs) have demonstrated remarkable capabilities, yet they occasionally exhibit sycophantic behavior, generating responses that align with or agree with a user's stated opinions or preferences, even when those opinions are incorrect or biased. This sycophantic tendency can undermine the trustworthiness and reliability of LLMs. This work proposes a novel approach to mitigate sycophancy in LLMs by fine-tuning them on a carefully curated dataset comprising prompts paired with sycophantic and non-sycophantic responses 1. Our method leverages Direct Preference Optimization (DPO), which optimizes LLMs to generate responses that align with the preferred (non-sycophantic) outputs without requiring explicit reward modeling. We develop a dataset of 1000 prompts with sycophantic and non-sycophantic responses to fine-tune LLMs. Our approach achieves an average reduction of 85% in persona-based tests and 84% in preference-driven tests, demonstrating significant mitigation of sycophantic behaviors. Our findings pave the way for more trustworthy and reliable language models that can provide objective and unbiased responses, aligning with human preferences while maintaining factual accuracy. © 2024 IEEE.

关键词： Finetuning Large Language Models

来源：评论

学校读者我要写书评

暂无评论

ACOVMD: Automatic COVID-19 misinformation detection in Twitter using self-trained semi-supervised hybrid deep learning model

引用

International Social Science Journal 2024年第252期74卷 713-730页

作者： Birunda, S. Selva Devi, R. Kanniga Muthukannan, M. Babu, M. Mahesh Department of Computer Science and Engineering School of Computing Kalasalingam Academy of Research and Education Tamil Nadu Srivilliputhur India School of Computer Science and Engineering Vellore Institute of Technology Chennai Tamil Nadu Kelambakkam India Department of Civil Engineering KCG College of Technology Tamil Nadu Chennai India

During the COVID-19 pandemic, online social networks are extensively utilized, more than ever before by 8.4%, resulting in the propagation of false information related to COVID-19. Despite the existence of many fake news detection models;annotation inconsistency, memory consumption, accurate and self-trained efficient algorithms for detecting the emerging COVID-19 misinformation tweets are still challenging. Hence, the main aim of this work is to come up with a self-trained semi-supervised model that accurately and automatically detects the reliability of emerging COVID-19 tweets without delay. In this work, COVID-19 tweet dataset is created in English Language from the period January 2020 to January 2022 as a ground truth database. Then self-trained semi-supervised hybrid deep learning model is proposed to train both supervised and unsupervised components simultaneously using the created dataset. The proposed model is self-trained repeatedly and the model gets updated to predict the reliability of upcoming COVID-19 tweets that differ from training tweets. We performed experiments multiple times by limiting the percentage amount of labelled tweets shown to the model, namely 80%, 50%, 40%, 30%, 20% and 10% labelled tweets, respectively. Experimental results show that the proposed model achieves 80.92% accuracy and 98.15% accuracy in the 10% and 80% label-seen experiments, respectively. This shows a clear rising trend in the performance curve. Therefore, this technique will be useful for effectively classifying voluminous amounts of emerging tweets generated as part of the COVID-19 infodemic. The proposed model may efficiently use a huge amount of unlabelled tweets and enhance the model's generalization performance. © 2023 John Wiley & Sons Ltd.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Secure UPI: Machine Learning-Driven Fraud Detection System for UPI Transactions 2

Secure UPI: Machine Learning-Driven Fraud Detection System f...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Rani, Rupa Alam, Adnan Javed, Abdul Ajay Kumar Garg Engineering College Ghaziabad Department of Computer Science & Engineering U.P. Ghaziabad India Banasthali Vidyapith Department of Computer Science & Engineering Rajasthan India Abes Institute of Technology Department of Computer Science & Engineering-Data Science U.P. Ghaziabad India

ISBN: (纸本)9798350371055

Secure UPI specializes in developing an advanced fraud detection gadget the usage of the effective XGBoost device getting to know set of rules to create an advanced fraud identity device. XGBoost is a properly-proper alternative for enhancing the precision of fraud detection fashions because of its reputation for dealing with tricky datasets with efficiency and its music record of success throughout multiple industries. In order to extract pertinent records, like transaction quantity, frequency, and place, our approach preprocesses UPI transaction facts. This article makes use of a labelled dataset to train the XGBoost version in order that we may also take gain of its sturdy prediction talents and capacity to handle imbalanced datasets. To help create a system that is less difficult to apprehend and use, feature importance evaluation is used to discover essential symptoms of feasible fraud. After training, the model is covered right into a real-time UPI transaction tracking device, in which it maintains an eye fixed out for any suspicious traits in incoming transactions. In order to lessen the results of fraudulent activity, the system is constructed with 98.2 % accuracy to send out instant notifications and take preventive steps. This challenge allows in improving UPI transaction security and advancing economic era are accomplished through demonstrating the performance of machine learning in fraud detection. © 2024 IEEE.

关键词： Dimensionality Reduction Fraud Detection Imbalanced Datasets Machine Learning PCA SMOTE UPI XGBoost

来源：评论

学校读者我要写书评

暂无评论

Detection of Mango Leaves Disease using CSUBW based Residual Attention with EfficientNet 2

Detection of Mango Leaves Disease using CSUBW based Residual...

引用

2nd International Conference on Distributed Computing and Optimization Techniques, ICDCOT 2024

作者： Kumar V, Sunil Jyothi, A.P. Nagamani, H.S. Sumanth, S. Nitte Meenakshi Institute of Technology Department of Artificial Intelligence and Machine Learning Bengaluru India Ramaiah University of Applied Sciences Faculty of Engineering and Technology Department of Cse Bengaluru India Department of Computer Science Bengaluru India Department of Computer Science and Computer Applications Kolar India

ISBN: (纸本)9798350382952

Mangoes are one of the most popular fruits worldwide due to their taste and nutrition. Plant diseases may reduce plant quantity and quality. Disease management to improve crop quality and productivity requires early plant illness detection. Automation of plant disease detection reduces large farm surveillance. Since plants feed us, leaf problems must be diagnosed quickly and accurately. Our advanced learning method automatically detects leaf disease in several mango plant varieties. This study creates the RAM-EfficientNet, a novel illness classification network. The RAM block in EfficientNet extracts additional features and handles slight leaf lesion changes. Transfer learning was used to train EfficientNet on ImageNet to overcome the mango leaf disease sample size issue. The Cat Swarm Updated Black Widow Model (CSUBW), a hybrid optimization model, changes RAM-EfficientNet's weights to improve classification accuracy. Black Widow Optimization Algorithm (BWO) and Cat Swarm Optimization Algorithm (CSO) are combined to create the hybrid optimization model. The model's performance is then assessed to confirm its efficacy. © 2024 IEEE.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Approach for Double Security with Watermarking and Chaos Based Encryption for e-Healthcare Systems 4

Approach for Double Security with Watermarking and Chaos Bas...

引用

4th IEEE International Conference on Electrical, computer, and Energy Technologies, ICECET 2024

作者： Sharma, Suvita Rani Singh, Birmohan Kaur, Manpreet Singh, Harpreet Sant Longowal Institute of Engineering and Technology Department of Computer Science and Engineering Punjab Sangrur148106 India Sant Longowal Institute of Engineering and Technology Department of Electrical and Instrumentation Engineering Punjab Sangrur148106 India

ISBN: (纸本)9798350395914

Medical data of the patients includes sensitive information and needs security while transmit over the internet. The double security approach based on watermarking and encryption method has been purposed in this work to secure the patient information consisting personal information, clinical information, and diagnostic records along with the medical images. The watermarking method is introduced on the basis of the least significant bit (LSB) method that conceals the patient information in the LSB bit of the image to hide the information. As, the LSB bit of the image is consider as least informative bit. Thus, it reduces the loss of data in the image. Then, the watermarked image is secured using the hybrid encryption method based on the Self Adaptive Bald Eagle Search (SABES) optimization algorithm, multiple chaotic maps, and circular shift. Chaotic maps are employed in the proposed approach to increase randomness in the encrypted image and chaotic maps' parameters are optimized employing the SABES optimization method to increase the chaotic maps efficiency. The propounded approach is tested on different types of medical images i.e., Grayscale, RGB, and hyperspectral images having different modalities and sizes. The performance of the propounded approach has been analyzed utilizing the Normalized Correlation (NC), Peak Signal-to-Noise Ratio (PSNR), entropy, histogram, Number of Changing Pixel Rate (NPCR), Structural Similarity Measure (SSIM), and Unified Averaged Changed Intensity (UACI) test. The proposed approach performance shows that the medical information secure from different types of attacks. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

PromptArt: AI-Powered Image Generation 8

PromptArt: AI-Powered Image Generation

引用

8th IEEE International Conference on Computing, Communication, Control and Automation, ICCUBEA 2024

作者： Patel, Jay Tale, Krishna Bidwe, Ranjeet Vasant Mishra, Sashikala Deshmukh, Ganesh Shinde, Sumit University Symbiosis Institute of Technology Pune Campus Pune412115 India Pimpri Chinchwad College of Engineering Department of Computer Engineering Pune India Itm Skills University Department of Computer Engineering Navi Mumbai India

ISBN: (纸本)9798350391770

Conventional picture creation requests noteworthy ability and labor, frequently constraining inventive expression to a select gather of capable specialists. This restrictiveness makes obstructions for amateurs and devotees, smothering differing qualities and advancement in visual craftsmanship. Whereas cutting-edge AI frameworks guarantee computerized picture amalgamation, they habitually battle to decipher nuanced printed prompts precisely, driving client disappointment when the coming about pictures don't meet desires. PromptArt addresses these challenges by leveraging Generative Ill-disposed Systems (GANs) to create high-quality pictures from literary depictions. This venture aims to democratize the inventive handle, engaging people of all skill levels to make personalized pictures easily. PromptArt's dual-stage GAN design first translates textual prompts into starting picture drafts and after that refines these drafts to improve quality and coherence, guaranteeing that the visual yields closely adjust with the client eagerly. With a user-friendly interface, PromptArt makes it simple for clients to input printed depictions and get high-quality visual representations. This openness changes the imaginative preparation, permitting more individuals to lock in in visual craftsmanship creation and contributing to a more different and inventive social scene. This paper investigates the engineering plan and execution of PromptArt, specifying its strategies and comes about. We moreover talk about the broader suggestions of democratizing picture creation, highlighting how PromptArt can cultivate more prominent imagination, collaboration, and social engagement, revolutionizing the way we make and connect with visual craftsmanship. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Removing Gaussian Image Noise Using Advanced Non-Local Means Filtering Techniques

Removing Gaussian Image Noise Using Advanced Non-Local Means...

引用

2023 International Conference on New Frontiers in Communication, Automation, Management and Security, ICCAMS 2023

作者： Suresh, A. Waris, Saiyed Faiayaz Mohankumar, B. Arumuga Arun, R. Saravanan, G. Vellore Institute of Technology School of Computer Science and Engineering Department of Computational Intelligence Vellore India Vignan's Foundation for Science Technology & Research Vadlamudi Department of Computer Science and Engineering Andhrapradesh Guntur India Vellore Institute of Technology School of Computer Science and Engineering Vellore India Veltech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering India

ISBN: (纸本)9798350317060

Our system is aimed at improving the application of image de-noising processes, even though the presence of noise has a negative impact on image quality. End-user applications suffer as a result of the focusing Gaussian noise. Introduce the Non-local means (NLM) filter to take advantage image similarities and analyse data preservation in de-noising. Even though the NLM approach eliminates some noises, like salt and pepper, an estimate weighted between the input images and edge preservation is still calculated. It could experience loss of visual structure, noise reduction, and blurring. This research shows the enhancement of the threshold level by combining stationary wavelet transform with NLM. This type of Advanced NLM method is effective at removing high noisy regions like Gaussian noise levels. The proposed ANLM method improves edge preservation and image structure similarities. This article uses some noisy images as input, then employs quantitative parameters to perform the measurement. Finally, this method demonstrates the effective removal noise and edge preservation of weighted refined structures. PSNR, COC, and MSE are the measuring parameters. © 2023 IEEE.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

Innovative Machine Learning Techniques for Accurate Detection of Bacterial Blight in Rice Agriculture 8

Innovative Machine Learning Techniques for Accurate Detectio...

引用

8th International Conference on Electronics, Communication and Aerospace technology, ICECA 2024

作者： Singh, Gurdeep Mittal, Sonam Chitkara University Institute of Engineering and Technology Chitkara University Department of Electronics and Communication Engineering Punjab India Chitkara University Institute of Engineering and Technology Chitkara University Department of Computer Science and Engineering Punjab India

ISBN: (纸本)9798350367904

Bacterial blight is a devastating disease that can significantly impact rice production. Early and accurate detection of the disease is crucial for effective management. This research proposes a deep learning-based approach, specifically utilizing Convolutional Neural Networks (CNNs), to identify and classify different severity levels of bacterial blight in rice plants. The proposed model involves a multi-step process, including data preprocessing, feature extraction, and classification. The CNN architecture is designed to capture the visual characteristics of infected plants, enabling accurate disease diagnosis. The model's performance is evaluated using metrics such as accuracy, precision, recall, and F1-score. By implementing this automated disease detection system, farmers can make timely decisions regarding disease management strategies, leading to improved crop yields and reduced economic losses. This research contributes to the advancement of precision agriculture and sustainable food production. © 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

A Modified Hexagonal Shaped Sub-6 GHz 5G Antenna in MIMO Configuration

A Modified Hexagonal Shaped Sub-6 GHz 5G Antenna in MIMO Con...

引用

2024 IEEE Microwaves, Antennas, and Propagation Conference, MAPCON 2024

作者： Porwal, Rahul Garg, Rahul Kumar Raj, Nikhil Nair, M.V. Deepak Department of Electronics and Communication Engineering The LNM Institute of Information Technology Jaipur India Department of Computer Science and Engineering University of Engineering and Management Jaipur India

ISBN: (纸本)9798350379693

A line-fed Modified Hexagon shaped 2-port planar antenna is presented for 5G application. The MIMO antenna is designed through combining the techniques of a modifying the radiator and defected ground structure. In the MIMO configuration, two replicas of the MIMO element are arranged orthogonal to ensure the pattern diversity. A cross shaped structure is integrated with the ground to enhance interelement isolation. The size and operating bandwidth of the MIMO antenna are 28×56 mm2 and 2.54 GHz to 5.46 GHz, respectively. This design advantageously offers a wide -15 dB bandwidth from 3.3 GHz to 5.0 GHz with a percentage bandwidth of 40.96% and a peak realized gain of 3.1 dBi at 3.3 GHz. Maximum isolation value of 33.55 dB is achieved while it is never less than 16.85 dB for the frequency of interest. The envelope correlation coefficient (ECC) © 2024 IEEE.

关键词： 5G mobile communication systems

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：