检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Li, Guozheng Xu, Zijie Shang, Ziyu Liu, Jiajun Ji, Ke Guo, Yikai School of Computer Science and Engineering Southeast University Bangladesh Beijing Institute of Computer Technology and Application China

Dialogue relation extraction (DRE) aims to extract relations between two arguments within a dialogue, which is more challenging than standard RE due to the higher person pronoun frequency and lower information density in dialogues. However, existing DRE methods still suffer from two serious issues: (1) hard to capture long and sparse multi-turn information, and (2) struggle to extract golden relations based on partial dialogues, which motivates us to discover more effective methods that can alleviate the above issues. We notice that the rise of large language models (LLMs) has sparked considerable interest in evaluating their performance across diverse tasks. To this end, we initially investigate the capabilities of different LLMs in DRE, considering both proprietary models and open-source models. Interestingly, we discover that LLMs significantly alleviate two issues in existing DRE methods. Generally, we have following findings: (1) scaling up model size substantially boosts the overall DRE performance and achieves exceptional results, tackling the difficulty of capturing long and sparse multi-turn information;(2) LLMs encounter with much smaller performance drop from entire dialogue setting to partial dialogue setting compared to existing methods;(3) LLMs deliver competitive or superior performances under both full-shot and few-shot settings compared to current state-of-the-art;(4) LLMs show modest performances on inverse relations but much stronger improvements on general relations, and they can handle dialogues of various lengths especially for longer sequences. © 2024, CC BY.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Improved Speaker-Emotion Voice Conversion with Limited Corpus based on Large Language Model and Pre-trained Model 6

Improved Speaker-Emotion Voice Conversion with Limited Corpu...

引用

6th International Conference on Mobile Computing and Sustainable Informatics, ICMCSI 2025

作者： Fanchang TangYaLing Yadav, Amit Sah, Abhishek Pratap Khan, Asif Department of Software Engineering Chengdu Neusoft University Chengdu China College of Engineering IT and Environment Charles Darwin University NT Australia School of Management Science and Engineering University of Electronic Science and Technology of China Chengdu China Department of Computer Application Integral University Lucknow India

ISBN: (纸本)9798331522667

Existing research on speaker and emotional voice conversion often focuses on separate tasks, neglecting their joint exploration. Furthermore, the limited availability of emotional corpora for target speakers poses a significant challenge for training robustness and generalized models. This paper proposes an improved scheme for speaker-emotion voice conversion with limited target speaker's emotional corpus, integrating a large language model and a pre-trained emotional speech synthesis model. It introduces several enhancements to enhance the quality of converted speech in terms of speaker similarity and emotional expressiveness. First, emotionally tagged text is generated using a large language model and emotional speech is synthesized from this text using a fine-tuned pre-trained emotional speech synthesis model. Then, a speaker-emotion voice conversion model is co-trained with both synthesized and real target emotional speech. Finally, the model is fine-tuned with the real target emotional speech to further boost the speaker and emotional similarity. © 2025 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Comparison of Machine Learning Techniques Based on Feature Reduction to Detect Diabetic Retinopathy 9

Comparison of Machine Learning Techniques Based on Feature R...

引用

9th International Conference on Advanced Computing and Communication Systems, ICACCS 2023

作者： Singh, Devendra Joshi, Deepika Dobhal, Dinesh C. Pant, Janmejay Pant, Himanshu Graphic Era Hill University Computer Science and Engineering Department Nainital India Amrapali Group of Institutes Faculty of Technology and Computer Application Haldwani India Graphic Era Deemed to Be University Computer Application Department Uttarakhad Dehradun India Graphic Era Hill University School of Computing Department Nainital India

ISBN: (纸本)9798350397376

Diabetic retinopathy is considered as an eye disorder condition that can affect persons who are diagnosed positive against diabetes. This condition causes damage to the retina's blood vessels. Diabetic retinopathy detection at an early stage may help patients protect their vision. This disease is often asymptomatic in the early stages but has a significant impact in the latter stages. The timely identification and intervention of diabetic retinopathy (DR) can greatly mitigate the potential for visual impairment. It takes time, effort, and money for ophthalmologists to manually diagnose DR retina fundus images. So there is a need for a screening system that can detect the disease in its early stages. There are various machine learning-based approaches available to detect diabetic retinopathy. In our proposed work first, we have to select the optimal features then we perform the comparative analysis with the help of various machine learning algorithms. We have used Random Forest, Logistic Regression, Gradient Boosting, Support Vector Machine (SVM), Naïve Bayes, and AdaBoost. In our experiment, we observed that Logistic Regression achieved the highest accuracy of 74.2 with optimal features. © 2023 IEEE.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

Vibration-induced bias error reduction using loop gain compensation for high-precision fiber optic gyroscopes

引用

Journal of Systems engineering and Electronics 2025年第1期36卷 224-232页

作者： CHEN Heyu QIN Xuexin XIE Huan KONG Linghai ZHENG Yue School of Instrumentation Science and Opto-Electronics Engineering Beihang UniversityBeijing 100191China Beijing Institute of Computer Technology and Application Beijing 100854China Hunan Shuaiwei Control Technology Co. LTDChangsha 421000China Beijing Institute of Control and Electronic Technology Beijing 100038China

Vibration-induced bias deviation,which is generated by intensity fluctuations and additional phase differences,is one of the vital errors for fiber optic gyroscopes(FOGs)operating in vibration environment and has severely restricted the applications of high-precision *** conventional methods for suppressing vibration-induced errors mostly concentrate on reinforcing the mechanical structure and optical path as well as the compensation under some specific operation parameters,which have very limited effects for high-precision FOGs maintaining performances under *** this work,a technique of suppressing the vibration-induced bias deviation through removing the part related to the varying gain from the rotation-rate output is put ***,the loop gain is extracted out by adding a gain-monitoring *** demodulating the loop gain and the rotation rate simultaneously under distinct frequencies and investigating their quantitative relationship,the vibrationinduced bias error is compensated without limiting the operating parameters or environments,like the applied modulation *** experimental results show that the proposed method has achieved the reduction of bias error from about 0.149°/h to0.014°/h during the random vibration with frequencies from20 Hz to 2000 *** technique provides a feasible route for enhancing the performances of high-precision FOGs heading towards high environmental adaptability.

关键词： vibration-induced error fiber optic gyroscope(FOG) loop gain

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis Scheme for Image-Text Comments based on Multimodal Deep Learning and Spatiotemporal Attention 6

Sentiment Analysis Scheme for Image-Text Comments based on M...

引用

6th International Conference on Mobile Computing and Sustainable Informatics, ICMCSI 2025

ISBN: (纸本)9798331522667

A sentiment analysis scheme for image and text comments based on multimodal deep learning and spatiotemporal attention is proposed to address the issues of incomplete spatiotemporal considerations, incomplete implementation details, and cutting-edge theoretical algorithms in graphic and textual sentiment analysis schemes. The proposed model has clear layering including data preprocessing layer, modal encoding layer, modal fusion layer, sentiment classification layer, loss function and optimizer, evaluation and feedback. The implementation details of each layer are introduced. The entire scheme model incorporates Multimodal Fusion Neural Network (MFNN) deep learning and spatiotemporal attention mechanism, which makes the scheme perform well in terms of security, robustness and performance, making up for the shortcomings of existing research schemes. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Utilizing Random Forests for High-Accuracy Classification in Medical Diagnostics 7

Utilizing Random Forests for High-Accuracy Classification in...

引用

7th International Conference on Contemporary Computing and Informatics, IC3I 2024

作者： Maindola, Meenakshi Al-Fatlawy, Ramy Riad Kumar, Rakesh Boob, Nandini Shirish Sreeja, S.P. Sirisha, N. Srivastava, Abhinav Graphic Era Deemed to be University Department of Computer Science & Engineering Dehradun India The Islamic university College of technical engineering Department of computers Techniques engineering Najaf Iraq Gla University Department of Computer Engineering & Applications Mathura India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Tamilnadu Chennai India New Horizon College of Engineering Master of Computer Application Bangalore India Mlr Institute of Technology Department of Computer Science and Engineering Telangana Hyderabad India Fs University Shikohabad Firozabad India

ISBN: (纸本)9798350350067

In recent years, medical diagnostics has increasingly relied on machine learning techniques to improve accuracy and efficiency. Among these, the Random Forest algorithm has emerged as a powerful tool for classification tasks due to its robustness and ability to handle complex datasets. This paper explores the application of Random Forests in medical diagnostics, specifically focusing on enhancing the accuracy of disease classification. By analyzing large, heterogeneous medical datasets, we demonstrate how Random Forests can outperform traditional classification algorithms, offering high precision in identifying and classifying diseases. Key features, such as decision tree ensemble methods and bootstrap aggregation, are utilized to minimize overfitting and improve model generalization. The research compares Random Forests to other machine learning models, such as Neural Networks and Support Vector Machines, and finds that Random Forests perform better in terms of dependability and classification accuracy. We further emphasise that the technique can scale to big datasets and that it can manage missing data. The findings show that Random Forests might be a useful tool for accurate and early diagnosis when incorporated into clinical procedures. The findings of this study highlight the promise of machine learning, and Random Forests in particular, to revolutionise healthcare diagnostics and enhance patient results. © 2024 IEEE.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

Traffic Prediction Analysis (TPA) Using Machine Learning Methodologies 6

Traffic Prediction Analysis (TPA) Using Machine Learning Met...

引用

6th International Conference on Contemporary Computing and Informatics, IC3I 2023

作者： Solanki, Mohit Rawat, Vandana Pandey, Monika Singh, Neelam Aswal, Shobha Graphic Era Deemed to Be University Department of Computer Application Dehradun India Graphic Era Deemed to Be University Department of Computer Science and Engineering Dehradun India School of Computing Graphic Era University Dehradun India

ISBN: (纸本)9798350304480

A traffic prediction system might help drivers in making better travel choices. Minimizing carbon emissions, reducing traffic congestion, and improving the effectiveness of traffic management to deliver such traffic flow data is the main aim of traffic flow prediction. Prediction of traffic is essential in a system of intelligent transportation. Effective traffic forecasting can help with route design, vehicle dispatching, and reducing traffic congestion. Due to its incredible capacity to predict traffic patterns, a branch of machine learning which is also called 'deep learning,'has recently attracted a lot of attention. In this study, a comparative evaluation has been done to predict traffic flow based on deep learning. To prepare the stacked auto encoder model, basic traffic flow features are included. The goal of this study is to conduct a thorough screening for deep earning -based traffic forecast algorithms from multiple angles. This study recommends a deep learning method to predict urban traffic that combines data from twitter messages with traffic and weather data. The predictive model uses a deep Bi-directional long- short-term memory (LSTM) Stacked Auto Encoder (SAE) architecture for multi-step traffic flow prediction that was trained from tweets, traffic, and weather datasets. To support study in this area, it also gathers and arranges frequently available public datasets. © 2023 IEEE.

关键词： Long-short Tem Memory (LSTM Stacked Auto Encoder (SAE etc

来源：评论

学校读者我要写书评

暂无评论

An end-to-end repair-based joint training framework for weakly supervised pavement crack segmentation

引用

Multimedia Tools and applications 2024年第16期84卷 1-19页

作者： Zhou, Hui Tao, Huanjie Duan, Qianyue Hu, Zhenwu Deng, Yishi School of Computer Science Northwestern Polytechnical University Xi’an710129 China School of Cyberspace Security Northwestern Polytechnical University Xi’an710129 China Engineering Research Center of Embedded System Integration Ministry of Education Xi’an710129 China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi’an710129 China

Weakly supervised pavement crack segmentation aims to assign each pixel of pavement surface images a category label (crack or non-crack) using limited annotation information. Most existing methods adopt the multi-stage approach based on the class activation map (CAM), which outputs from a pre-trained crack classification model, to generate the pixel-level pseudo labels, and then train a segmentation model using these pseudo labels. However, these methods have two notable drawbacks: (i) CAM focuses primarily on the most discriminative regions of the image, potentially resulting in an incomplete estimation of crack regions, and (ii) the process of obtaining pixel-level pseudo labels from CAM is complex and may introduce noise. To avoid these issues, we propose an end-to-end repair-based joint training framework for weakly supervised pavement crack segmentation. Specifically, to better guide the model to pay more attention to the repair process of crack regions, we introduce the segmentation loss, which assigns different weights for crack regions and background regions, through the Residual Project Network. Additionally, to alleviate the impact of the domain gap between synthetic and real datasets, we introduce the consistency loss through the joint training paradigm of unpaired data that incorporates both crack and non-crack images into the training process simultaneously. Furthermore, to enable the model to effectively extract local features and reduce the model's computational complexity, we introduce the Lightweight Multi-head Self-attention module. The experiment results on the CRACK500 Dataset, the CFD Dataset, and the AEL Dataset demonstrate that our model achieves better performance than some existing weakly supervised pavement segmentation methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Repair

来源：评论

学校读者我要写书评

暂无评论

Enhancing Rule-Based Methods for Pulmonary Effusion Diagnosis 15

Enhancing Rule-Based Methods for Pulmonary Effusion Diagnosi...

引用

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

作者： Kharbas, Vikas Kumar Raghu, N. Choudhury, Priyabati Vivekananda Global University Department of Computer Science & Application Jaipur India School of Engineering and Technology Department of Electrical and Electronics Engineering Karnataka Bangalore India Arka Jain University Department of Pharmacy Jharkhand Jamshedpur India

ISBN: (纸本)9798350370249

The diagnosis of pulmonary effusion is a difficult assignment for scientific practitioners because of the excessive variance in the presentation and manifestation of the circumstance. Maximum current diagnosis structures rely upon the use of either sensory/visible remark of chest radiographs (CRs) by radiologists or handcrafting statistical features through area experts with the use of computer-aided analysis (CAD) tools. However, those methods regularly need better generalization and are constrained via their reliance on conventional heuristic algorithms, as well as the mistake-inclined technique of guide identification of features. This paper explores the capacity of the usage of rule-based total methods as better diagnostic structures for pulmonary effusion. Particularly, a rule-based total device primarily based on supervised mastering-based total fashions is proposed as an alternative to traditional feature identification and model development strategies. The proposed device includes an output layer and an enter layer, which includes photograph features and radiological descriptors. Experiments have been conducted on the LIDC-IDRI dataset, and the proposed system outperformed traditional feature-based approaches in terms of both accuracy and specificity. Furthermore, the use of rule-based methods for diagnosis of pulmonary effusion provides a more reliable diagnosis process, which may reduce the radiological review times and improve the accuracy of non-radiological diagnosis. © 2024 IEEE.

关键词： Conducted Diagnosis Effusion Identification Manifestation Pulmonary Radiographs

来源：评论

学校读者我要写书评

暂无评论

Evaluating the Impact of Different Features on the Efficiency of Emotion Recognition Algorithms in Text 2

Evaluating the Impact of Different Features on the Efficienc...

引用

2nd International Conference on Artificial Intelligence and Machine Learning applications, AIMLA 2024

作者： Sharma, Sunil Ezhilarasan, G. Ghosh, Rajkumari Vivekananda Global University Department of Computer Science & Application Jaipur India School of Engineering and Technology Department of Electrical and Electronics Engineering Karnataka Bangalore India ARKA JAIN University Department of English Jharkhand Jamshedpur India

ISBN: (纸本)9798350349221

The effectiveness today's emotion recognition algorithms in texts has become increasingly more critical in current years today's its programs in supporting users gauge actual-time reactions from text-based totally conversations. On this paper, we advise a fixed state-of-the-art experiments to assess the effect that different functions, along with syntactic and semantic functions, have at the efficiency contemporary emotion reputation algorithms in textual content. We use a manually annotated dataset from the automatic have an effect on recognition in text (AART) task to create characteristic units consultant of various ranges latest functions. Those feature units are then used to assess the overall performance latest emotion recognition algorithms - Recursive Neural Networks and Convolutional Neural Networks. The results display that incorporating syntactic and semantic functions can improve the accuracy present day emotion popularity algorithms at a charge ultra-modern 6-10% while as compared with the baseline overall performance. From the findings modern this test, we conclude that incorporating syntactic and semantic features can drastically enhance the overall performance trendy emotion popularity algorithms in text-based totally conversations. © 2024 IEEE.

关键词： Semantics

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：