检索结果-内蒙古大学图书馆

Proceedings of the 2023 7th international conference on Natural Language processing and Information Retrieval

作者： Jingdong Wang Yongjia Guo School of Computer Science Northeast Electric Power University China

ISBN: (纸本)9798400709227

As a crucial first step in the process of constructing knowledge graph, the accuracy of named entity recognition determines the construction effect of the final graph. However, at present, Chinese named entity recognition methods still have many problems, such as long training time and lower accuracy. Hence, we come up with a BERT-BILSTM-ACRF entity recognition method that combines the “self-attention” mechanism. To begin with, Bert model is selected as the embedding layer, the text is vectorized, and the character position in-formation is obtained through the bidirectional Long Short-Term Memory network. Secondly, the internal relationship of the character sequence is further searched through the self-attention mechanism, and finally the final optimal sequence is decoded by the conditional random field model. To check the effectiveness of the BERT-BILSTM-ACRF model, the model is applied to the data set of the university course textbook” Da-ta Structure”, and the result reaches 98.97%F1 value and 98.14%accuracy, which has good experimental results and certain practical value.

关键词： Knowledge graph Named entity recognition Natural language processing Self-attention mechanism

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Supervised and Semi-supervised Training Algorithms of Restricted Boltzmann Machines on Biological Data

A Comparison of Supervised and Semi-supervised Training Algo...

引用

international Symposium on Computational Intelligence and Informatics

作者： Gergely Pap László Tóth Institute of Informatics University of Szeged Szeged Hungary

ISBN: (数字)9781728156255

ISBN: (纸本)9781728156262

Most devices applied nowadays in biological experiments produce large quantities of data. However, due to the cost of human expertise, only a small subset of these data can be examined or labeled by humans, while a huge part of the extracted information remains unlabeled. therefore, interest in machine learning methods that can automatically make use of unlabeled data (sometimes in parallel with the annotated data parts) have grown steadily in the past years. In this study we show that one approach to handle the issue stated above could be the application of Restricted Boltzmann Machines (RBMs) in a semi-supervised machine learning configuration. RBMs can extract relevant information from the unlabeled data, and use it to improve the classification accuracy achieved by learning only on the labeled data set. We compare semi-supervised classification RBMs with a (deep) Artificial Neural Network and a Support Vector Machine to examine the improvements originating from the use of the unannotated data in three different, biologically inspired data sets. Our results show that even in this modern era of deep learning, semi-supervised classifiers are a viable option in these areas of science where annotating data is a time-consuming and resource-heavy task.

关键词： semi-supervised learning Restricted Boltzmann Machines classification biological data Boltzmann machines Biological data Semi-supervised Learning Taxonomy Support Vector Network SUPERVISING

来源：评论

学校读者我要写书评

暂无评论

Acceleration of blender cycles path-tracing engine using intel many integrated core architecture

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2015年 9339卷 86-97页

作者： Jaroš, Milan Říha, Lubomír Strakoš, Petr Karásek, Tomáš Vašatová, Alena Jarošová, Marta Kozubek, Tomáš IT4Innovations VšB-Technical University of Ostrava Ostrava Czech Republic Department of Applied Mathematics VšB-Technical University of Ostrava Ostrava Czech Republic

ISBN: (纸本)9783319243689

this paper describes the acceleration of the most computationally intensive kernels of the Blender rendering engine, Blender Cycles, using Intel Many Integrated Core architecture (MIC). the proposed parallelization, which uses OpenMP technology, also improves the performance of the rendering engine when running on multi-core CPUs and multi-socket servers. Although the GPU acceleration is already implemented in Cycles, its functionality is limited. Our proposed implementation for MIC architecture contains all features of the engine with improved performance. the paper presents performance evaluation for three architectures: multi-socket server, server with MIC (Intel Xeon Phi 5100p) accelerator and server with GPU accelerator (NVIDIA Tesla K20m). © IFIP international Federation for Information processing 2015.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Research on Signal Denoising Method Based on Symplectic Geometric Mode Decomposition and Savitzky-Golay Filtering

Research on Signal Denoising Method Based on Symplectic Geom...

引用

IEEE international conference on Electronic Information and Communication Technology (ICEICT)

作者： Xuelong Xu Jin Zhao Bo Wu Xianliang Wu Jincheng Ni Ping Wu Key Lab of Intelligent Computing & Signal Processing Ministry of Education Anhui University Anhui University Hefei China

ISBN: (数字)9798350384437

ISBN: (纸本)9798350384444

the paper proposes a noise reduction algorithm based on symplectic Geometric Mode Decomposition (SGMD) and Savitzky-Golay (SG)filtering to address the issue of noise interference during signal transmission. Firstly, the noisy signal is decomposed using symplectic geometric mode decomposition to obtain several symplectic geometric components. the component with less noise is then selected and superimposed based on its energy entropy. Finally, the reconstructed signal undergoes denoising through SG filtering. the proposed method is applied to conduct comparison experiments on noise reduction for linear frequency modulation signals and multicomponent periodic signals. the results demonstrate a significant noise reduction effect of the proposed method, which holds great significance for signal analysis and research on noise reduction.

关键词： Smoothing methods Noise reduction Noise Interference Information filters Robustness Signal denoising

来源：评论

学校读者我要写书评

暂无评论

Modeling of Dust Deposition Impact on PV/T Hybrid Collector Performances

Modeling of Dust Deposition Impact on PV/T Hybrid Collector ...

引用

international Renewable and Sustainable Energy conference (IRSEC)

作者： Chaimae El Fouas Oussama El Manssouri Bekkay Hajji Loubna Bouselham Giuseppe Marco Tina Antonio Gagliano Renewable Energy Embedded System and Data Processing Laboratory National School of Applied Sciences Mohamed First University Oujda Morocco University of Catania Catania Italy

ISBN: (数字)9781728151526

ISBN: (纸本)9781728151533

Dust deposition on the front of photovoltaic /thermal collectors presents greatest issues in some regions (desertic regions). It's considered as key parameter affecting much more collector's performances. In the present work, a particular attention has been paid to dust deposition effect on the electrical and thermal performances of photovoltaic /thermal collector PV/T.A model of sheet-and-tubes PV/ T collector is developed and validated with experimental data available in literature. the obtained results show that the dust significantly affects the electrical and thermal efficiencies. An increase of dust deposition of 8g/m2 causes an electrical and a thermal output reduction respectively of 28% and 21%. Better electrical and thermal performances have been noted for PV/ T without dust. Moreover, a good agreement has been observed between obtained results and the presented results available in literature.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exploiting Combination of Spectral Features in Light Weight Neural Network for ABE on Limited Data

Exploiting Combination of Spectral Features in Light Weight ...

引用

international conference on Signal processing and Information Security (ICSPIS)

作者： Ritujoy Biswas Karan Nathwani Mohammed Krini Electrical Engineering Indian Institute of Technology Jammu Jammu India Laboratory for Signal Processing Aschaffenburg University of Applied Sciences Aschaffenburg Germany

ISBN: (数字)9798350368673

ISBN: (纸本)9798350368680

this paper proposes an approach to extend the bandwidth of 4 KHz speech signals upto 8 KHz using a simple feed-forward neural network (FFNN). We employ the commonly used Short-Time Fourier Transform (STFT) values as features and boost its resolution by incorporating Mel Frequency Cepstral Coefficients (MFCCs). Apart from this feature-level fusion, a layer-level fusion is also examined to demonstrate the expansion in bandwidth. this is an advantage over traditional artificial bandwidth extension (ABE) methods using neural nets, which are heavily data-driven and require high model complexity. the proposed approach exploits the combination of relevant spectral features to avoid the requirement of highly complex architectures. this is because the combined features capture more intricate features relating the 4 KHz signal to the 8 KHz one. this facilitates the deployment of the technique on resourceconstrained devices, even with limited data. these reconstructed high pass (8 KHz) audio signals are perceptually better than the original 4 KHz bandwidth signals. Additionally, the performance of the proposed approach exceeds that of state-of-the-art methods in the literature. this approach has potential applications in enhancing the quality by extending the bandwidth of audio signals in telecommunication systems and also finds application in cochlear implant (CI) users.

关键词： Performance evaluation Cochlear implants Fourier transforms Information security Bandwidth Telecommunications Feedforward neural networks Complexity theory Mel frequency cepstral coefficient Signal resolution

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Based Cloud Microphysics Parameterization Scheme for Numerical Weather Prediction with 1DD-CNN

Deep Learning Based Cloud Microphysics Parameterization Sche...

引用

international conference on Artificial Intelligence and Big Data (ICAIBD)

作者： Ting Shu Xiyuan Zhou Yuanfu Xie Huan Zhao Xunlai Chen Rui Wang Shenzhen Institute of Meteorological Innovation Shenzhen China The Chinese University of Hong Kong Shenzhen Shenzhen China The Hong Kong Polytechnic University Hong Kong China Shenzhen Meteorological Bureau / Shenzhen Key laboratory of severe weather in South China Shenzhen China

ISBN: (数字)9798350385106

ISBN: (纸本)9798350385113

Machine learning methods, represented by deep learning, are rapidly changing and enhancing the process and results of the Numerical Weather Prediction (NWP) model. To better explore the application potential of the machine learning method in the Microphysical Parameterization Schemes (MPS) for NWP model, this study introduces a deep learning based cloud microphysical processes scheme. the scheme includes two steps: data processing and machine learning. the data process step in our scheme is to prepare data for the machine learning method, where the data is extracted and refined from NWP outputs. Subsequent to this process, a one-dimensional dense convolutional neural network (1DD-CNN) is applied to rigorously analyze the data and boost the predictive precision and computational speed of cloud microphysical processes. Empirical results validate the scheme's efficacy in advancing meteorological predictive capabilities.

关键词： Deep learning Meteorological radar Clouds Computational modeling Weather forecasting Predictive models Data processing

来源：评论

学校读者我要写书评

暂无评论

Trustworthy and Robust Machine Learning for Multimedia: Challenges and Perspectives

Trustworthy and Robust Machine Learning for Multimedia: Chal...

引用

IEEE conference on Multimedia Information processing and Retrieval (MIPR)

作者： Katsuaki Nakano Michael Zuzak Cory Merkel Alexander C. Loui Department of Computer Engineering Rochester Institute of Technology Rochester NY USA

ISBN: (数字)9798350351422

ISBN: (纸本)9798350351439

Multimedia applications for machine learning mod-els are characterized by the fusion of multiple modalities of data. In this work, we highlight the trust and robustness challenges of machine learning that arises from data fusion. To do so, we present three case studies demonstrating how multimedia applications exacerbate existing challenges of trustworthy and robust machine learning. For the first case study, we investigate the impact of fusion depth on the robustness of multi-modal machine learning models, observing that model architecture could impact robustness. For the second case study, we investigate the impact of fusion modality on the robustness of multi-modal machine learning models, observing that fusion models are only as robust as their most susceptible modality. For the third case study, we explore the impact of weight quantization techniques on the robustness of multi-modal models, observing the need for modality-based quantization schemes. through these case studies, we hope to shed light on the unique trust and security challenges that arise in machine learning models when applied in multimedia applications and offer insights to fortify such systems in real-world scenarios.

关键词： Quantization (signal) Multimedia systems Data integration Machine learning Information processing Robustness Security

来源：评论

学校读者我要写书评

暂无评论

A Real-time AD-Census Stereo Matching based on FPGA

A Real-time AD-Census Stereo Matching based on FPGA

引用

international conference on Control, Automation and Systems ( ICCAS)

作者： Yongseok Lee Seong-Bin Choi Eunchong Lee Sang-Seol Lee Sung-Joon Jang Intelligent Image Processing Research Center Korea Electronics Technology Institute Seoul Korea

In this paper, we present an environment for the acquisition of a flight image of drone and design an stereo matching accelerator based on FPGA(Field Programmable Gate Array) to acquire depth information at real-time from acquired images. the acquisition of the flight images of drone uses a stereo camera (ZED) and Jetson TX2 board environment. the obtained binocular image can acquire depth information in real time through a stereo matching accelerator. Stereo matching operations have large computational complexity and require hardware acceleration. In this paper, we design an FPGA based stereo matching accelerator. Accelerators implemented AD-Census stereo matching algorithm and Box Filtering cost aggregation function at high speed. Census cost calculation method uses 9×7 filter, and cost aggregation function uses 15×15 filter. In order to reduce the internal memory usage, a cost aggregation function with parallel operation is implemented and a dynamic fixed point operation method is applied in order to reduce the amount of computation block usage. As a result, the 480×270 resolution image on the KU040 FPGA board showed 41fps performance at 100MHz clock operating speed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Influence of 16k-30kHz Sinusoidal Voltage on the Partial Discharge and Breakdown Characteristics of parallel Wires at Low Pressure

Influence of 16k-30kHz Sinusoidal Voltage on the Partial Dis...

引用

China international Electrical and Energy conference (CIEEC)

作者： Qiaopo Xiong Gen Long Tianwei Wang Yaoxun Zhao She Chen WMCRI Hubei Provincial Key Laboratory of Low Frequency Electromagnetic Communication Technology Wuhan China College of Electrical and Information Engineering Hunan University Changsha China

ISBN: (数字)9798350359558

ISBN: (纸本)9798350359565

Due to the requirements of certain special communication frequency bands (Very Low Frequency, VLF) for high-altitude platforms, the internal insulation design of power source faces challenges posed by low air pressure and voltage frequencies in the range of 3–30 kHz. Presently, limited research exists on the partial discharge and breakdown characteristics within the VLF band, particularly concerning the impact on discharge characteristics under low pressure conditions. In this study, we established an experimental platform where a 16k-30kHz VLF sinusoidal voltage was applied to a parallel conductor gap. We measured the Partial Discharge Inception Voltage (PDIV) and breakdown voltage within the pressure range of 19.4kPa to 57.8kPa. the results indicated a notable decrease in PDIV with voltage frequency for a 10 cm air gap. For a 2 cm gap distance, the breakdown voltage under 50Hz voltage was marginally lower than that under VLF frequency at low pressure, although they exhibited similarity at atmospheric pressure. As the gap distance increased to 3cm, the breakdown voltages under the two voltage types became comparable under low pressure conditions. Upon reaching atmospheric pressure, the breakdown voltage under VLF frequency was markedly lower than that under power frequency conditions.

关键词： Partial discharges Partial discharge measurement Breakdown voltage Insulation Voltage measurement Electric breakdown Wires

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：