检索结果-内蒙古大学图书馆

A novel graph kernel algorithm for improving the effect of text classification

Computer Speech & Language 2026年 95卷

作者： Fan Yang Tan Zhu Jing Huang Zhilin Huang Guoqi Xie College of Computer and Information Engineering Central South University of Forestry and Technology Changsha Hunan 41004 PR China School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan Hunan 411201 PR China Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University Changsha Hunan 410082 PR China

Text classification is an important topic in natural language processing. In recent years, both graph kernel methods and deep learning methods have been widely employed in text classification tasks. However, previous graph kernel algorithms focused too much on the graph structure itself, such as the shortest path subgraph,while focusing limited attention to the information of the text itself. Previous deep learning methods have often resulted in substantial utilization of computational resources. Therefore,we propose a new graph kernel algorithm to address the disadvantages. First,we extract the textual information of the document using the term weighting scheme. Second,we collect the structural information on the document graph. Third, graph kernel is used for similarity measurement for text classification. We compared eight baseline methods on three experimental datasets, including traditional deep learning methods and graph-based classification methods, and tested our algorithm on multiple indicators. The experimental results demonstrate that our algorithm outperforms other baseline methods in terms of accuracy. Furthermore, it achieves a minimum reduction of 69% in memory consumption and a minimum decrease of 23% in runtime. Furthermore, as we decrease the percentage of training data, our algorithm continues to achieve superior results compared to other deep learning methods. The excellent experimental results show that our algorithm can improve the efficiency of text classification tasks and reduce the occupation of computer resources under the premise of ensuring high accuracy.

关键词： Document similarity measure Graph kernel Machine learning Term weighting Text classification

来源：评论

学校读者我要写书评

暂无评论

Energy optimization on dynamic multiple functions automotive cyber-physical systems 21

Energy optimization on dynamic multiple functions automotive...

引用

21st IEEE International Conference on High Performance computing and Communications, 17th IEEE International Conference on Smart City and 5th IEEE International Conference on Data Science and Systems, HPCC/SmartCity/DSS 2019

作者： Liu, Yan Xie, Guoqi Jin, Linlin Li, Renfa Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University Changsha Hunan China

ISBN: (纸本)9781728120584

A new generation of automotive electronic systems is a typical cyber-physical systems (CPS) based on an integrated architecture, in which multiple functions can be executed on one electronic control unit (ECU) and one function can be assigned to different ECU. Heterogeneity and dynamics are essential characteristics of automotive CPS (ACPS). Considering heterogeneity and dynamics of ACPS, optimizing functional safety and energy consumption in system design is still a challenge. In this paper, we focus on the collaborative optimization of functional safety and energy consumption with global method in the context of prioritizing functional safety. Considering the requirements of functional safety, we propose a deadline driven processor merging for multiple dynamic functions (DPMMDF) algorithm to maximizes the number of functions that meet deadlines when a function arrives dynamically from the perspective of global optimization. Further, we use a global energy saving for multiple dynamic functions (GESMDF) algorithm, combined with the integrated architecture features, to optimize the energy consumption of ACPS by selecting multiple candidate executable ECUs for a task. Experimental results show that combination of the proposed two algorithms can effectively reduce deadline missed ratio (DMR), and minimize energy consumption compared with existing methods. © 2019 IEEE.

关键词： Global optimization

来源：评论

学校读者我要写书评

暂无评论

A Sentiment Classification Model Based on Multiple Multi-classifier Systems 1

引用

5th International Conference on Artificial Intelligence and Security, ICAIS 2019

作者： Yang, Kehua Liao, Chenglin Zhang, Wei College of Computer Science and Electronic Engineering Key Laboratory for Embedded and Network Computing of Hunan Province Hunan University ChangshaHunan410082 China

ISBN: (数字)9783030242657

ISBN: (纸本)9783030242640

With the fast development of the Internet, especially the development of the Web technique, more and more people express their views on the Internet. The content of these views involve various aspects of our life, such as books, movies, musics, policies, commodities, technical services, etc. Since these views are usually subjective and possess great potential commercial and social value, the study of sentiment analysis is followed. At the early stage of sentiment analysis, researchers used lexicon based method to solve sentiment analysis problem. With the improvement in machine learning and Natural Language Processing technology, machine learning based method are becoming more and more popular in the field of sentiment analysis. In this study, we propose an ensemble sentiment classification model based on three multi-classifier systems. The proposed model uses majority voting as an ensemble method of multiple multi-classifier systems. The effectiveness of the classification model is verified on five public datasets. Experimental results show that the proposed model is superior to traditional sentiment classification methods and simple multi-classifier systems in the performance of classification accuracy. By using the sentiment classification model proposed in this paper, the accuracy improvement in different domains also show that our proposed model has certain generalization ability. © 2019, Springer Nature Switzerland AG.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Optimizing Word Embedding for Fine-Grained Sentiment Analysis 1

引用

5th International Conference on Artificial Intelligence and Security, ICAIS 2019

作者： Zhang, Wei Zhang, Yue Yang, Kehua Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University Changsha410082 China

ISBN: (数字)9783030242657

ISBN: (纸本)9783030242640

Word embeddings have been extensively used for various Natural Language Processing tasks. However, word vectors trained based on corpus context information fail to distinguish words with the same context but different semantics, which may lead to a similar word vector with opposite semantic terms. This will affect some Natural Language Processing tasks, such as fine-grained sentiment analysis tasks. In this paper, a new word vectors optimization model is proposed. This model can be applied to any pre-trained word vectors. Within a certain range, it can make the opposite semantic words away from each other, and the same semantic words are close to each other. The experimental results show that our model can improve the traditional word embedding in the fine-grained emotional analysis task of Chinese Weibo, and the optimized word vector using our model outperforms the unoptimized word vector. © 2019, Springer Nature Switzerland AG.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Real-time depth estimation with an optimized encoder-decoder architecture on embedded devices 21

Real-time depth estimation with an optimized encoder-decoder...

引用

作者： Tu, Xiaohan Xu, Cheng Liu, Siping Xie, Guoqi Li, Renfa Key Laboratory for Embedded and Network Computing of Hunan Province China College of Computer Science and Electronic Engineering Hunan University Changsha410082 China

ISBN: (纸本)9781728120584

As a strategy to obtain dense depth maps from a single image, sparse depths, or the fusion of both (RGBd), depth estimation receives much attention. Usually relying on high-performance workstations, existing depth prediction methods, are not particularly suitable for embedded devices. Moreover, the accuracy of depth estimation needs to be improved. Towards real-time and reliable depth prediction on portable embedded platforms, we propose an encoder-decoder architecture (EDA). Specifically, these are the three contributions of this work. (1) We present a light-weight encoder, which takes three modalities of data as inputs, namely, RGB images, sparse depths, or RGBd. The encoder is efficient to extract features on portable devices. (2) A novel decoder is designed to obtain accurate and high-resolution depth maps. (3) We use the existing TVM to deploy, compile, and optimize EDA on embedded platforms, for further improvements in the computational speed and power of depth estimation. In real scenarios, the optimized EDA achieves remarkable results with low latency. Through extensive experimental evaluation on two public datasets, we show that the optimized EDA outperforms the state of the arts in accuracy and computational resources. To the best of our knowledge, this is the first paper that provides an efficient encoder, learnable decoder, and end-to-end optimized encoder-decoder architecture for precise depth prediction. © 2019 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

TMI: Two-dimensional maintainability index for automotive software maintainability measurement

引用

Journal of Systems Architecture 2025年 166卷

作者： Xie, Jiao Yang, Jialing Zhao, Sirong Lei, Jianmei Fan, Kaiwei Xie, Guoqi Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University Hunan Changsha410082 China Center of Convergence of Automobile and Cyberspace Research Institute of Hunan University in Chongqing Chongqing401120 China China Automotive Engineering Research Institute Co Ltd Electronic Communication and Software Evaluation Research Center Chongqing401122 China

The mass adoption of autonomous driving and Advanced Driver-Assistance System (ADAS) technologies increases the complexity and maintenance cost of automotive-embedded software during model-based development. Through model-based development tools (e.g., MATLAB/Simulink), the code is automatically generated by architectural transformation;however, unreasonable software architecture design could make the final generated code be difficult to maintain. Currently, both Microsoft Maintainability Index (MMI) and Bosch Maintainability Index (BMI) are used to measure the maintainability of code, but neither software architecture. We establish a four-layer maintainability index model by analyzing the characteristics of automotive software architecture, software quality model, and industry-related standards. We then propose a Two-dimensional Maintainability Index (TMI), meaning the code and architecture dimensions are simultaneously considered. This is the first work of maintainability measurement to combine code and architecture dimensions while ensuring low redundancy. TMI utilizes nine index metrics to guide architecture refactoring and code optimization, significantly improving software maintainability. © 2025 Elsevier B.V.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

A deep hashing method based on attention module for image retrieval 13

A deep hashing method based on attention module for image re...

引用

13th International Conference on Intelligent Computation Technology and Automation, ICICTA 2020

作者： Long, Jun Wei, Xiangxiang Qi, Qianqian Wang, Ye School of Computer Science and Engineering Central South University Changsha410083 China Network Resources Management and Trust Evaluation Key Laboratory of Hunan Province Changsha410083 China

ISBN: (纸本)9781728186665

Hashing retrieval technique has received widespread attention since its low storage consumption and fast retrieval speed. Image retrieval based on deep hashing method uses image features with high-level semantic information and discriminative expression capabilities. However, there are still some limitations: (1) Compared with the traditional feature vector, the length of the learned hash codes is short, and each bit of the code is -1 or 1, the discrimination capability of feature representation is limited;(2) the existing deep hash algorithms cannot directly learn the discrete hash codes using sign activation function, therefore, they use relaxation-based scheme to learn the hash codes, resulting in a large quantization error. In order to solve the above problems, a deep hashing retrieval algorithm incorporate with attention model is proposed. This algorithm embeds spatial and channel attention models in the feature extraction network, and uses a novel activation function in the hash layer. In addition, a novel triple loss function is further proposed to improve the learning capability of the network. Finally, experimental results on the two benchmark databases, i.e., CIFAR-10 and MINIST datasets, show the effectiveness and superiority compared with the baselines. © 2020 IEEE.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

ACWB: Access-characteristic write buffer management scheme for flash memory 21

ACWB: Access-characteristic write buffer management scheme f...

引用

作者： Zhu, Lei Liu, Yan Zhang, Zaimei Key Laboratory for Embedded and Network Computing of Hunan Province College of Computer Science and Electronic Engineering Hunan University Changsha Hunan410082 China School of Economics and Management Changsha University of Science and Technology Changsha Hunan410076 China

ISBN: (纸本)9781728120584

Compared with traditional hard disk drives (HDD), solid state drives (SDD) have distinct characteristics, such as efficient continuous access and random reading capability. Using of SSD is greatly limited due to the out-of-place update and endurance problem. Many flash-aware buffer schemes have been proposed to reduce write and erase operations. However, a good buffer management scheme should reduce the effects of random writes and suit access patterns. In this paper, we propose an access characteristic write buffer management scheme (ACWB) on flash memory that integrates write buffer and cost regulation based on access characteristic. Experimental results show that ACWB increases the response time by 35%, reduces erase times by 25%, and improves the hit ratio by 27% on the average compared with current flash-aware write buffer schemes. © 2019 IEEE.

关键词： Flash memory

来源：评论

学校读者我要写书评

暂无评论

Scholars influence evaluation based on time series heterogeneous network

Scholars influence evaluation based on time series heterogen...

引用

International Conference on Measuring Technology and Mechatronics Automation, ICMTMA

作者： Ye Wang Zhen Ding Xiang Xiang Wei Jun Long The School of Computer Science and Engineering Central South University Changsha China Network Resources Management and Trust Evaluation Key Laboratory of Hunan Province Changsha China

In recent years, with the rapid development of network and data mining, multi-source heterogeneous data related to academics has also shown an exponential growth. As an important research field of academic big data, the objectivity of academic influence evaluation methods and rationality plays an important role in the evaluation of scientific and technological talents, academic groups, and national scientific research strength. Traditional methods are generally measured based on the amount of papers published by scholars and the total frequency of citations. Although these methods are simple to calculate the influence of scholar, they ignore the importance of the cooperative network of scholar and the evolution of academic networks over time. In order to solve the above limitations, this paper constructs a derivation of scholars' paper relationship network and scholars' cooperation network, taking into account the importance of time in the network, from the perspectives of the journal where the scholar publishes papers, the location of the author, and the number of citations of the paper. In this paper, a method is designed to calculate the academic influence of the scholars, making the calculation result more reasonable.

关键词： Computer science Mechatronics Automation network topology Time series analysis Cooperative systems Big Data

来源：评论

学校读者我要写书评

暂无评论

BGRA-Net: Boundary-Guided and Region-Aware Convolutional Neural network for the Segmentation of Breast Ultrasound Images

BGRA-Net: Boundary-Guided and Region-Aware Convolutional Neu...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Xiang Zhang Xuanya Li Kai Hu Xieping Gao Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Xiangtan China Baidu Inc. Beijing China Key Laboratory of Medical Imaging and Artifical Intelligence of Hunan Province Xiangnan University Chenzhou China Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing Hunan Normal University Changsha China

ISBN: (纸本)9781665429825

In this paper, we propose a novel convolutional neural network based on boundary-guided and region-aware (BGRA-Net) for breast tumor segmentation in ultrasound images. In particular, in the encoding stage, we propose a boundary-guided module (BGM) to guide the learning of boundary features in the decoding stage by explicitly strengthening the extracted boundary information. Meanwhile, in the decoding stage, we propose a region-aware module (RAM) to integrate different levels of detailed and semantic features to improve the comprehensive representation of tumor regional features. Besides, a scale-adaptive module (SAM) is further proposed to capture the characteristics of tumors with different sizes between the encoding and decoding stages. To evaluate the effectiveness of our BGRA-Net, we conduct extensive experiments on the UDIAT dataset and compare it with eight state-of-the-art methods. The experimental results show that our BGRA-Net outperforms the state-of-the-art methods and can achieve accurate segmentation of breast tumors with ambiguous boundaries.

关键词： Deep learning Image segmentation Breast tumors Ultrasonic imaging Semantics Feature extraction Visual effects

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：