检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第3期6卷 589-603页

作者： Qiu, Jian-Qiang Zhang, Chun-Yang Chen, C. L. Philip Fuzhou University College of Computer and Data Science Fuzhou350108 China South China University of Technology School of Computer Science and Engineering Guangzhou510006 China

Pretrained language models (PLMs) have shown remarkable performance on question answering (QA) tasks, but they usually require fine-tuning (FT) that depends on a substantial quantity of QA pairs. Therefore, improving the performance of PLMs in scenarios with only a small number of training examples, also known as a few-shot setting, is of great practical significance. Current mitigation strategies for the few-shot QA task largely rely on pretraining a QA task-specific language model from scratch, overlooking the potential of foundational PLMs to generate QA pairs, particularly in the few-shot setting. To address this issue, we propose a prompt-based QA data augmentation method aimed at automating the creation of high-quality QA pairs. It employs the PFT method, adapting the question generation process of PLMs to the few-shot setting. Additionally, we introduce a dynamic text filling training strategy. This strategy simulates the progressive human learning process, thereby alleviating overfitting of PLMs in the few-shot setting and enhancing their reasoning capability to tackle complex questions. Extensive experiments demonstrate that the proposed method outperforms existing approaches across various few-shot configurations. 2691-4581 © 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://***/publications/rights/*** for more information.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection Model Using Chaotic MAP for Network Coding Enabled Mobile Small Cells

引用

computers, Materials & Continua 2024年第3期78卷 3151-3176页

作者： Chanumolu Kiran Kumar Nandhakumar Ramachandran School of Computer Science and Engineering VIT-AP UniversityAmaravatiAndhra Pradesh522241India

Wireless Network security management is difficult because of the ever-increasing number of wireless network malfunctions,vulnerabilities,and *** security systems,such as Intrusion Detection Systems(IDS),are essential due to the limitations of simpler security measures,such as cryptography and *** to their compact nature and low energy reserves,wireless networks present a significant challenge for security *** features of small cells can cause threats to the *** Coding(NC)enabled small cells are vulnerable to various types of *** attacks and performing secure“peer”to“peer”data transmission is a challenging task in small *** to the low power and memory requirements of the proposed model,it is well suited to use with constrained small *** attacker cannot change the contents of data and generate a new Hashed Homomorphic Message Authentication Code(HHMAC)hash between transmissions since the HMAC function is generated using the shared *** this research,a chaotic sequence mapping based low overhead 1D Improved Logistic Map is used to secure“peer”to“peer”data transmission model using lightweight H-MAC(1D-LM-P2P-LHHMAC)is proposed with accurate intrusion *** proposed model is evaluated with the traditional models by considering various evaluation metrics like Vector Set Generation Accuracy Levels,Key Pair Generation Time Levels,Chaotic Map Accuracy Levels,Intrusion Detection Accuracy Levels,and the results represent that the proposed model performance in chaotic map accuracy level is 98%and intrusion detection is 98.2%.The proposed model is compared with the traditional models and the results represent that the proposed model secure data transmission levels are high.

关键词： Network coding small cells data transmission intrusion detection model hashed message authentication code chaotic sequence mapping secure transmission

来源：评论

学校读者我要写书评

暂无评论

Leveraging dual-blockchain collaboration for logistics supply chain supervision

IET Blockchain

引用

IET Blockchain 2024年第S1期4卷 570-585页

作者： Lin, Licheng Pan, Senshan Jing, Pujie Song, Xiangmei School of Computer Science and Telecommunication Engineering Jiangsu University Jiangsu Zhenjiang China

Logistics supply chain (LSC), a chain structure that integrates and coordinates all logistics transactions, has become an essential component of the modern logistics industry. By using blockchain, trusted logistics services enable participants to effectively record and track transactions during the logistics process. Current blockchain-based LSC features distributed structure and data privacy requirements, hindering the supervision of logistics transactions. Leveraging emerging dual-blockchain architecture to separate logistics transactions from supervision is a promising direction. However, the dual-blockchain collaboration restricts supervision due to its cross-chain privacy and efficiency. To address these issues, a logistics supply chain supervision scheme based on dual-blockchain collaboration (DBC) is proposed. First, an independent supervision blockchain is constructed to balance the contradiction between distributed structure and supervision requirements. Second, two mechanisms are designed to enhance the privacy and performance of collaborative supervision. The hybrid access control mechanism enables fine-grained supervision for different participants, and the aggregated transaction verification method supports efficient collaboration for logistics transactions. Security analysis and performance evaluation demonstrate the feasibility of DBC in enhancing the security and supervision of logistics data on the dual-blockchain architecture. Experimental results show that the cross-chain supervision overhead of DBC is reduced to (Formula presented.) of the baseline schemes. © 2024 The Author(s). IET Blockchain published by John Wiley & Sons Ltd on behalf of The Institution of engineering and Technology.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

RLeU-Net: Segmentation of blood vessels in retinal fundus images for Diabetic Retinopathy Screening

引用

Multimedia Tools and Applications 2025年第9期84卷 6113-6134页

作者： Addanki, Sathyavani Sumathi, D. School of Computer Science and Engineering VIT-AP University Andhra Pradesh Amaravati India

Diabetic Retinopathy (DR) is a primitive cause of blindness in diabetic patients. There are chances for DR in certain cases, due to the damage in the retinal vascular area. It may lead to blindness or loss of vision if DR is not detected. At the final stage, there is a threat of loss of vision. During the initial stage of this chronic disease, it has been observed that no symptoms are required to determine the loss of vision. Hence, there is a need for better and early diagnosis of the blood vessels to prevent vision loss in diabetic patients. The primary task of early diagnosis is to carry out a proper segmentation process. But, segmenting vasculture from retinal photographs is a vital laborious effort. Therefore, blood vessel segmentation of fundus images has grown in favor among researchers. There are many existing methods to process bio-medical retinal images, and most of them are computationally intensive and difficult to deploy in real-time. The objective of this work is to develop a novel and efficient deep learning framework to perform the segmentation with the U-Net as a primary module. The U-Net architecture has been modified so that the model would be lightweight and the model is cost efficient in terms of time and space complexity. The proposed method, DRLeU- Net, is evaluated using a publicly available color retinal fundus images dataset. The model was trained using the k-fold cross-validation technique. The proposed method shows a promising results of 0.988 AUC (Area Under the Curve) and 0.94 IOU (Intersection over Union when compared with the existing *** robust and efficient deep learning framework has been deployed to do the segmentation process efficiently and it has evauated with the expremental results by comparing with the existing methodologies. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

A Robust Decryption Technique Using Letter Frequency Analysis for Short Monoalphabetic Substitution Ciphers

引用

Journal of Computing science and engineering 2024年第3期18卷 144-151页

作者： Kang, Dayeong Lee, Jiyeon School of Computer Science and Engineering Kyungpook National University Daegu Korea Republic of

Substitution ciphers—where characters are systematically replaced with different ones—have represented a vital tool in safeguarding sensitive information for centuries, eventually laying the foundation for modern encryption techniques. Despite their large key space, substitution ciphers are susceptible to frequency analysis that leverages common English letter patterns. While existing research has suggested certain methods that can improve decryption accuracy using ngram frequencies, these methods face difficulties when used with short ciphertexts due to incomplete letter distribution representation. The present study examines the limitations of current frequency analysis in decrypting short ciphertexts, with the results revealing that deterministic bigram approaches can reduce accuracy in certain cases. To address this shortcoming, we introduce a novel algorithm that uses randomized index selection based on letter distribution to generate multiple candidate keys. We also present a word-level key guessing method using these candidates that maps prominent English words to uncover a secret key. The results of tests with 200 ciphertexts of varying lengths showed an average decryption accuracy of 84.1% for 200-character ciphertexts, an improvement of 147.1% over existing methods. In experiments without dictionary-based decryption, an accuracy of 77.6% was achieved with a decryption time of approximately 0.27 seconds, which is a reasonable completion time. Altogether, these results highlight the efficiency and practicality of our approach for decrypting short ciphertexts. © 2024. The Korean Institute of Information Scientists and Engineers

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

Enhancing demand forecasting through combination of anomaly detection and continuous improvement

引用

Soft Computing 2025年第2期29卷 1243-1258页

作者： Jahani, Meysam Zojaji, Zahra Raji, Fatemeh Faculty of Computer Engineering University of Isfahan Isfahan Iran School of Computer Science and Informatics De Montfort University Leicester United Kingdom

Demand forecasting has emerged as a crucial element in supply chain management. It is essential to identify anomalous data and continuously improve the forecasting model with new data. However, existing literature fails to comprehensively cover both aspects of anomaly detection and continuous improvement in demand forecasting. This study proposes an enhanced model to improve accuracy in the demand forecasting. The proposed model introduces a novel data handling method that incorporates an anomaly detection autoencoder, improved with anomaly correction mechanisms. The data handling approach simultaneously detects data anomalies, distinguishes between expected and unexpected anomalies, and corrects anomalous data, ensuring cleaner input for demand forecasting. Then, the proposed model employs a long short-term memory architecture for demand forecasting, enhanced with a continuous improvement method. Thus, the model not only forecasts demand but also retrains the model when the anomaly data surpasses the predetermined threshold, thereby improving the accuracy of forecasting. The results show that the proposed model outperforms other models in detecting data anomalies, achieving an average precision-recall of 0.922, a receiver operating characteristic value of 0.739, and a significance level of less than 0.05. Finally, the model exhibits superior performance in demand forecasting, with average mean squared error, root mean squared error, and mean absolute error values of 33.167, 4.347, and 1.509, respectively, all with a significance level of less than 0.05. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Demand response

来源：评论

学校读者我要写书评

暂无评论

Efficient Multimodal Selection for Retrieval in Knowledge-based Visual Question Answering

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 5195-5207页

作者： Luo, Linyin Lai, Hanjiang Pan, Yan Yin, Jian Sun Yat-Sen University School of Computer Science and Engineering China School of Artificial Intelligence China

Retrieval plays an important role in knowledge-based visual question answering (KB-VQA), which relies on external knowledge to answer questions related to an image. However, not all information in the external knowledge is beneficial in retrieval, e.g., the knowledge that is only semantically similar to the query but is not useful for question answering. To improve the effectiveness and efficiency of retrieval, in this paper, we propose efficient multimodal selection to filter out irrelevant information and increase the retriever performance for KB-VQA. First, to exclude most irrelevant knowledge from the large external knowledge, multimodal selection uses a query-aware sample selection method, which uses the pretrained answer generator's prediction to obtain better positive and negative training samples to help retrievers distinguish knowledge that is semantically relevant to the multimodal query. Then, question-aware visual feature selection is proposed to select the distinguishable visual information related to the question: where cross-attention to questions and images is proposed to obtain question-aware visual features. These visual features are used to perform fine-grained multimodal retrieval within the small set to obtain the final top-related knowledge. The experimental results show that the proposed approach achieves state-of-the-art retrieval performance on the OK-VQA and FVQA datasets, indicating the effectiveness of our selection strategy for retrieval. © 1991-2012 IEEE.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

2P3FL:A Novel Approach for Privacy Preserving in Financial Sectors Using Flower Federated Learning

引用

computer Modeling in engineering & sciences 2024年第8期140卷 2035-2051页

作者： Sandeep Dasari Rajesh Kaluri School of Computer Science Engineering and Information Systems Vellore Institute of TechnologyVellore632014India

The increasing data pool in finance sectors forces machine learning(ML)to step into new *** data has significant financial implications and is *** users data from several organizations for various banking services may result in various intrusions and privacy *** a result,this study employs federated learning(FL)using a flower paradigm to preserve each organization’s privacy while collaborating to build a robust shared global ***,diverse data distributions in the collaborative training process might result in inadequate model learning and a lack of *** address this issue,the present paper proposes the imple-mentation of Federated Averaging(FedAvg)and Federated Proximal(FedProx)methods in the flower framework,which take advantage of the data locality while training and guaranteeing global *** improves the privacy of the local *** analysis used the credit card and Canadian Institute for Cybersecurity Intrusion Detection Evaluation(CICIDS)***,recall,and accuracy as performance indicators to show the efficacy of the proposed strategy using FedAvg and *** experimental findings suggest that the proposed approach helps to safely use banking data from diverse sources to enhance customer banking services by obtaining accuracy of 99.55%and 83.72%for FedAvg and 99.57%,and 84.63%for FedProx.

关键词： Federated learning FedAvg FedProx flower framework privacy preservation financial sectors

来源：评论

学校读者我要写书评

暂无评论

Enhancing the Well-Being of Seniors: A Teachable Agent for Ikigai

引用

International Journal of Crowd science 2024年第2期8卷 71-76页

作者： Chen, Ping Yu, Xinjia Lim, Su Fang Shen, Zhiqi School of Computer Science and Engineering Nanyang Technological University Singapore639798 Singapore

Ikigai, a Japanese term often referred to as "purpose in life" or "the feeling that life is worthwhile", is a commonly used indicator of well-being in studies focused on the elderly population in Japan. It can be identified within the different realms like family, work, and friendship. However, as individuals age, their sense of ikigai is prone to diminish. Teachable agents (TAs) have been employed in educational settings for decades to facilitate learning by teaching students. Previous research has indicated that TAs can have a positive impact on self-esteem which is an important aspect of ikigai. TAs may hold the potential to address issues related to declining ikigai that are often associated with aging. By engaging in the act of teaching the agent, elderly individuals may experience a sense of being needed, consequently enhancing their self-esteem and potentially leading to an improvement in their ikigai levels. In this paper, we introduce a TA designed around the concept of ikigai, aiming at assisting the elderly in maintaining a high level of ikigai to support healthy aging. Based on a user study conducted using the phenomenographical approach, we demonstrate the effectiveness of our proposed TA design. © The author(s) 2024.

关键词： Sociology Education Oral communication Knowledge graphs Aging Older adults Statistics

来源：评论

学校读者我要写书评

暂无评论

An enhanced AMBTC for color image compression using color palette

引用

Multimedia Tools and Applications 2024年第11期83卷 31783-31803页

作者： Xiong, Lizhi Zhang, Mengtao Yang, Ching-Nung Kim, Cheonshik School of Computer Science Nanjing University of Information Science and Technology Nanjing China Department of Computer Science and Information Engineering National Dong Hwa University Hualien Taiwan Department of Computer Engineering Sejong University Seoul Korea Republic of

Digital image has been used in various fields as an essential carrier. Many color images have been constantly produced since their more realistic description, which takes up much storage space and network bandwidth. Thus, color image compression has become an essential key technology. Absolute Moment Block Truncation Coding (AMBTC) has been widely studied as one of the classical image compression methods. However, in the existing methods, the visual quality of the reconstructed images and the compression rate are all relatively low. Therefore, this paper proposes an enhanced AMBTC for color image compression using a color palette. In the proposed method, the K-means clustering algorithm is utilized for training the image's palette pattern. The color palette obtained by K-mean will be more suitable for reconstructing this image than the standard color palette, and the visual quality will be higher. The six clustered central pixels are matched with the palette through a color difference formula, and the obtained index values are used as the quantization levels. Huffman coding is used to build a bitmap to achieve a higher compression rate, that is, a lower bit rate. At last, a block of a color image can be represented by six index values and a bitmap. Experimental results and theoretical analysis demonstrate that the proposed method has better visual quality and bit rate than similar schemes. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Color

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：