检索结果-内蒙古大学图书馆

20th International computer Conference on Wavelet Active Media technology and Information Processing, ICCWAMTIP 2023

作者： Liang, Tian School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (纸本)9798350318982

Source-data free domain adaptation (SFDA) is from unsupervised domain adaptation (UDA) and applied to the special situations in reality that the source domain data is not accessible. To this end, a common method is self-supervision with marking pseudo labels on the data in target domain. While the method is simple and effective, error hard labels degenerate the final model seriously. In this paper, we introduce the latent dirichlet allocation (LDA) algorithm to enhance the predictive ability of the target model. We consider all the sample features within a neighborhood follow the same distribution. The prediction from the target model is viewed as a prior distribution of the sample. And we improve the prediction with pseudo-labels of samples in its neighborhood with the help of LDA. Compared with the baseline SHOT-IM, the experiment results show that our method is improved 1.3% and 1.6% on datasets Office-31 and Office-Home respectively. © 2023 IEEE.

关键词： Statistics

来源：评论

学校读者我要写书评

暂无评论

Improving Speaker Verification Back-End with Graph Neural Networks

引用

Journal of Shanghai Jiaotong University (science) 2025年 1-9页

作者： Chen, Jinfeng Fang, Zhihua He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

Currently, research on speaker verification tasks is primarily concentrated on enhancing deep speaker models to extract high-quality speaker embeddings. Nevertheless, this speaker embeddings can be regarded as potential graph structures. Therefore, this paper introduces a graph neural network-based speaker verification back-end approach, treating the speaker embeddings derived from the front-end as graph structures and leveraging graph neural networks to uncover the relationships among embeddings, thereby achieving superior-quality speaker embeddings. In addition, we propose a group updating method to solve the problem of memory overflow when the number of nodes is excessive. Extensive experiments and ablation studies were carried out on the VoxCeleb dataset, and the experimental outcomes validate the effectiveness of our proposed graph neural network speaker back-end in significantly boosting the performance of speaker verification systems. (Figure presented.) © Shanghai Jiao Tong University 2025.

关键词： Graph embeddings

来源：评论

学校读者我要写书评

暂无评论

Machine learning-aided scoring of synthesis difficulties for designer chromosomes

引用

science China(Life sciences) 2023年第7期66卷 1615-1625页

作者： Yan Zheng Kai Song Ze-Xiong Xie Ming-Zhe Han Fei Guo Ying-Jin Yuan Frontiers Science Center for Synthetic Biology and Key Laboratory of Systems Bioengineering Ministry of EducationTianjin UniversityTianjin300072China School of Chemical Engineering and Technology Tianjin UniversityTianjin300072China School of Computer Science and Engineering Central South UniversityChangsha410083China

Designer chromosomes are artificially synthesized ***,these chromosomes have numerous applications ranging from medical research to the development of ***,some chromosome fragments can interfere with the chemical synthesis of designer chromosomes and eventually limit the widespread use of this *** address this issue,this study aimed to develop an interpretable machine learning framework to predict and quantify the synthesis difficulties of designer chromosomes in *** the use of this framework,six key sequence features leading to synthesis difficulties were identified,and an e Xtreme Gradient Boosting model was established to integrate these *** predictive model achieved high-quality performance with an AUC of 0.895 in cross-validation and an AUC of 0.885 on an independent test *** on these results,the synthesis difficulty index(S-index)was proposed as a means of scoring and interpreting synthesis difficulties of chromosomes from prokaryotes to *** findings of this study emphasize the significant variability in synthesis difficulties between chromosomes and demonstrate the potential of the proposed model to predict and mitigate these difficulties through the optimization of the synthesis process and genome rewriting.

关键词： synthetic biology machine learning artificial chromosome chemical synthesis

来源：评论

学校读者我要写书评

暂无评论

A Classification Method for Depression Based on MRI 20

A Classification Method for Depression Based on MRI

引用

20th International computer Conference on Wavelet Active Media technology and Information Processing, ICCWAMTIP 2023

作者： Zhou, Chunwen Zeng, Feng Xing, Jianchuan School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (纸本)9798350318982

Depression is a serious threat to human health, and the difficulty of diagnosis and stigmatized by society make it difficult to theat. More than 350 million people worldwide suffer from depression. Therefore, earlier and more accurate diagnosis can save patients' lives. Most depression detection methods such as deep learning methods and manual questionnaires, perform poor interpretability or need long detection time. Therefore, our designed a depression diagnosis method based on Magnetic resonance imaging data. Firstly, we preprocessed the image to obtain data with high signal-to-noise ratio. Secondly, we extracted structural brain networks and functional brain networks from the data as input features. Finally, we used a classification method base on support vector machines and light gradient boosting machine algorithm to obtain classification results. © 2023 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Enhanced Speech Emotion and Gender Recognition Using Hybrid CNN-LSTM Models 4

Enhanced Speech Emotion and Gender Recognition Using Hybrid ...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Priyadharshini, S. Sundar, G. Naveen Narmadha, D. Thangavel, Senthil Kumar Ajibesin, Adeyemi Abel Kolandapalayam Shanmugam, Selvanayaki Karunya Institute of Technology and Sciences Department of Computer Science and Engineering Coimbatore India Amrita School of Computing Department of Computer Science and Engineering Coimbatore India Cape Peninsula University of Technology Faculty of Informatics and Design Department of Information Technology South Africa Ashland University Department of Mathematics and Computer Science United States

ISBN: (纸本)9798331523923

The current paper proposes a new approach for peripheral speech emotion analysis and gender estimation incorporating the best machine learning architectures such as CNNs and LSTMs. Its correct depiction of emotions and the variation of the gendered speech to indicate manifold signal-to-noise ratio make the approach accommodate the difficulty inherent in recognizing emotions from the recording of voices as well as predicting the gender of a speaker. It is then able to ally with preprocessing and collection of a vast range of different speech samples, feature extraction of pitch, tone and spectral data. These features serve as input to the CNNs and LSTMs: LSTMs work well at temporal regular features and the CNNs work well in feature extraction by location (spatio feature). The approach is tested on the RAVDESS and IEMOCAP datasets and establishes a near perfect match to prior approaches with reference to recognition accuracy of emotion analysis and gender prediction. Thus, applying the identified dual method approach, which could be described by such characteristics as highly accurate data pre-preprocessing, clear indication of the features of interest, a different CNN-LSTM model joint training, it offers a reasonably good platform to gain certain insights into such characteristics of the speech as our experiment demonstrates. This work offers satisfactory configurations on concerns such as human computer relation, customer relation and mental health surveillance to suggest that other applications could be achieved in future. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Fuzzy based Fall Risk Prediction in Older Adults 3

Fuzzy based Fall Risk Prediction in Older Adults

引用

3rd International Conference on Artificial Intelligence For Internet of Things, AIIoT 2024

作者： Mohan, Deepika Al-Hamid, Duaa Zuhair Chong, Peter Han Joo Gutierrez, Jairo Li, Hui Auckland University of Technology Department of Electrical and Electronic Engineering Auckland New Zealand Auckland University of Technology Department of Computer Science and Software Engineering Auckland New Zealand Peking University School of Electronic & Computer Engineering Shenzhen China

ISBN: (纸本)9798350372120

The global elderly population is increasing rapidly, leading to a rise in chronic illnesses and co-existing conditions, which in turn results in higher healthcare expenses. Accidental falls are among the leading causes of injury-related deaths in elderly individuals. This study aims to create a real-time monitoring system using vital signs to foresee a future fall in older adults by identifying abnormalities through continuous monitoring. The proposed fall prediction technique, employing the Fuzzy-based Fall Prediction Algorithm, utilizes Fuzzy rules to learn and execute tasks. The obtained results are then classified according to different levels of predicted risk indicators. The developed model is tested using data from older adults sourced from a public repository and compared with the results of the theoretical evaluation. The simulated outcomes demonstrate that the proposed algorithm achieves 96% accuracy, 93.75% sensitivity, and 100% specificity. Utilizing these advancements in the proposed heterogeneous technology allows for the early prediction of falls in the elderly and can potentially save lives. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A Review of Driver Fatigue Detection Based on Facial Expression Recognition 20

A Review of Driver Fatigue Detection Based on Facial Express...

引用

20th International computer Conference on Wavelet Active Media technology and Information Processing, ICCWAMTIP 2023

作者： Jia, Chen Jianping, Li Changrun, Cui Lixi, Chen School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (纸本)9798350318982

Fatigue driving is one of the main causes of traffic accidents. Under fatigue, the driver's reaction time increases, and they cannot take timely remedial measures in emergencies, which leads to the occurrence of traffic accidents or worsens their severity. Real-time fatigue detection during driving is highly effective in preventing accidents caused by fatigue driving. The current fatigue detection methods can be roughly divided into three categories: (1) detection methods based on driver's physiological characteristics;(2) detection methods based on vehicle driving state;(3) detection methods based on driver's facial features. Because the first two methods have low robustness, the third method holds the most research significance. This paper analyzes several methods for detecting fatigue based on facial features. The text briefly describes the method without facial expression features, the method based on facial expression features, and the process of multi-feature fusion with facial expression features. It also summarizes and analyzes the methods and technologies used in the key steps of the process. The future research is also being anticipated. © 2023 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Enhanced CNN-DCT Steganography: Deep Learning-Based Image Steganography Over Cloud

引用

SN computer science 2024年第4期5卷 1-13页

作者： Ahmad, Shahnawaz Ogala, Justin Onyarin Ikpotokin, Festus Arif, Mohd. Ahmad, Javed Mehfuz, Shabana School of Computer Science Engineering and Technology Bennett University Uttar Pradesh Greater Noida 201310 India Department of Cyber Security University of Delta Agbor Nigeria Department of Computer Science Ambrose Alli University Edo State Ekpoma Nigeria Department of CSE School of Computing Science and Engineering Galgotias University Uttar Pradesh Greater Noida 203201 India Department of Computer Science and Engineering Sharda School of Engineering and Technology Sharda University Uttar Pradesh Greater Noida 201310 India Department of Electrical Engineering Jamia Millia Islamia New Delhi 110025 India

Image steganography plays a pivotal role in secure data communication and confidentiality protection, particularly in cloud-based environments. In this study, we propose a novel hybrid approach, CNN-DCT Steganography, which combines the power of convolutional neural networks (CNNs) and discrete cosine transform (DCT) for efficient and secure data hiding within images over cloud storage. The proposed method capitalizes on the robust feature extraction capabilities of CNNs and the spatial frequency domain transformation of DCT to achieve imperceptible embedding and enhanced data-hiding capacity. In the proposed CNN-DCT Steganography approach, the cover image undergoes a two-step process. First, feature extraction using a deep CNN enables the selection of appropriate regions for data embedding, ensuring minimal visual distortions. Next, the selected regions are subjected to the DCT-based steganography technique, where secret data is seamlessly embedded into the image, rendering it visually indistinguishable from the original. To evaluate the effectiveness of our approach, extensive experiments are conducted using a diverse dataset comprising 500 high-resolution images. Comparative analysis with existing steganography methods demonstrates the superiority of the proposed CNN-DCT Steganography approach. The results showcase higher data hiding capacity, superior visual quality with an MSE of 112.5, steganalysis resistance with a false positive rate of 2.1%, and accurate data retrieval with a bit error rate of 0.028. Furthermore, the proposed method exhibits robustness against common image transformations, ensuring the integrity of the concealed data even under various modifications. Moreover, the computational efficiency of our approach is demonstrated by a competitive execution time of 2.3 s, making it feasible for real-world cloud-based applications. The combination of deep learning techniques and DCT-based steganography ensures a balance between security and visual qual

关键词： Cloud storage CNN Data hiding DCT Deep learning Image steganography Visual fidelity

来源：评论

学校读者我要写书评

暂无评论

Latent Dirichlet Allocation for Topic Modeling and Intelligent Document Classification 2nd

Latent Dirichlet Allocation for Topic Modeling and Intelli...

引用

2nd International Conference on Innovations in Data Analytics, ICIDA 2023

作者： Chatterjee, Rajdeep Mukherjee, Chandan Chatterjee, Siddhartha Nath, Biswaroop School of Computer Engineering KIIT Deemed to be University Bhubaneswar751024 India School of Engineering and Technology The Assam Kaziranga University Jorhat785006 India Department of Computer Science and Engineering College of Engineering and Management Kolaghat West Bengal Purba Medinipu721171 India

ISBN: (纸本)9789819749270

Document classification plays a pivotal role in facilitating faster and more intelligent information retrieval. This paper specifically focuses on document image classification, utilizing optical character recognition (OCR) to transform it into a natural language processing (NLP) problem. Our research aims to enhance the accuracy of popular models by incorporating Latent Dirichlet Allocation (LDA), a text mining and topic modeling technique based on probabilistic distributions. By integrating LDA, we aim to leverage its capabilities for filtering the OCR-generated textual data. This filtering process enables us to extract important features and improve the performance of our models. In our study, we employed several models, in which the Long Short-Term Memory (LSTM) model, a widely used deep learning architecture, to tackle the document image classification task achieved the most promising results, with the LSTM achieving a notable accuracy rate of 92%. To establish the robustness of our findings, we conduct extensive testing on a separate, larger test dataset, which further validates the accuracy and reliability of our proposed approach. Real-life test scenarios are also considered, wherein our LSTM model successfully identifies different types of documents. The successful application of our model in real-world scenarios highlights its potential for practical implementation. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Fast Approximate Pattern Counting for Static Graphs 3

Fast Approximate Pattern Counting for Static Graphs

引用

3rd International Conference on Frontiers of Electronics, Information and Computation Technologies, ICFEICT 2023

作者： Xue, Ruini Wang, Yijun Chao, Lingwei Tian, Wenhong University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu610031 China

ISBN: (纸本)9798350302356

Pattern counting is a fundamental computational graph mining task. Recently, extensive research has been conducted on approximate pattern counting, but existing algorithms are not scalable, especially for large-scale static graphs. To address this challenge, we propose FAPC-A fast and efficient method for approximate pattern counting by exploiting the fact that the distribution of pattern numbers to degrees follows the power-law function as well. Unlike existing sampling processes, our approach can quickly fit the formula coefficients and calculate directly the pattern frequency. The prototype of FAPC has been evaluated against multiple public datasets and experimental results demonstrate that it outperforms current approaches by up to 50x speed. © 2023 IEEE.

关键词： graph pattern mining pattern counting power law distribution

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：