检索结果-内蒙古大学图书馆

37th Chinese Control Conference, CCC 2018

作者： Jixuan, Yang Qin, Zhang Peng, Yang Junqing, Sun Tianjin University of Technology Tianjin Key Laboratory of Intelligent Computing and Novel Software Technology Tianjin300384 China

ISBN: (纸本)9789881563941

The Laser Induced Breakdown Spectroscopy (LIBS) technology can be used to detect the elements in the alloy steel in real time. Quantitative analysis method of the traditional LIBS technology mainly has the calibration method and calibration free method, but there are two shortcomings: low prediction accuracy and over fitting. Random Forest Regression (RFR) algorithm can be used for classification and regression, can effectively avoid 'overfitting' phenomenon. Therefore, in this paper, we combine the random forest regression algorithm with laser induced breakdown spectroscopy applied to the detection of the concentration of alloy steel elements in the metallurgy industry. At the same time, compared with partial least squares method based on the LIBS, the results show that the random forest algorithm combined with the LIBS technology has the higher prediction accuracy, lower root mean square error and better robustness. © 2018 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Laser induced breakdown spectroscopy

来源：评论

学校读者我要写书评

暂无评论

The improvement of HDFS authentication model based on token push mechanism 18

The improvement of HDFS authentication model based on token ...

引用

2018 International Conference on Big Data and computing, ICBDC 2018

作者： Wang, Guiyuan Ning, Hongyun Tianjin Key Laboratory of Intelligence Computing and New Software Technology Tianjin China

ISBN: (纸本)9781450364263

In order to solve the problems of single point overload, repeated authentication, replay attack and time synchronization in the security authentication of HDFS, an improved Kerberos protocol based on token push mechanism is proposed. First of all, aiming at the problem of single point overload and repeated authentication, a three-stage access mechanism based on Agent is introduced. The KDC is responsible for the visitor's first login to HDFS for the first time. In the second phase, the visitor accesses the DataNode again through HDFS, the agent generates cross-node tokens and pushes to all involved DataNodes. In the third phase, visitor directly accesses DataNode. DataNode uses cross-node tokens to authenticate visitors. This mechanism divides users' login into three phases, which lightens the load of KDC authentication server. Secondly, reduce the number of NameNode and KDC authentication through the token push mechanism. Finally, in order to enhance the security of the authentication process and prevent replay attacks, a new parameter T-nonce is introduced in the authentication of Client, NameNode and DataNodes. This parameter is obtained by hash processing the timestamp and IP address. The timestamp and IP address are combined in the parameter generation function to ensure its real-time and uniqueness so that unauthorized users cannot implement replay attacks. This paper analyzes the security and efficiency of a single-time authentication both in the original model and the improved one, and the results show that the improved protocol can effectively improve the security and efficiency of HDFS authentication. © 2018 Association for computing Machinery.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Microblog text clustering based on BK-means algorithm 8

Microblog text clustering based on BK-means algorithm

引用

2018 8th International Workshop on Computer Science and Engineering, WCSE 2018

作者： Li, Qianru Mo, Xiuliang Wang, Chundong Tianjin Intelligent Computing and Software New Technology Key Laboratory School of Computer Science and Engineering Tianjin university of technology Tianjin300384 China

ISBN: (纸本)9789811178610

In recent years, the increasing popularity of social media such as WeChat and Weibo has facilitated the communication among people. However, due to the characteristics like large scale, fast propagation, low quality and diverse modalities of social short texts, the short text clustering faces the challenge of sparse features, high dimension and noise interference. The traditional clustering method based on vector space model is not good for short text data processing. With the improvement of K-means algorithm, this paper proposes a short-text clustering algorithm named BK-means which alleviates the effect of data sparseness. Firstly, we preprocess the wordset by means of word segmentation, stop-of-word and other operations, then extract the biterm using the BTM to model the document, and get the document-topic, the topic-word distribution matrix. Finally, we use the proposed BK-means algorithm to cluster short texts of documents represented by vectors. Experiments on the short text data of Sina Weibo have proved that the short text clustering algorithm based on BK-means is superior to the traditional one, and both the F-measure and the purity are improved. © Proceedings of 2018 the 8th International Workshop on Computer Science and Engineering, WCSE 2018. All rights reserved.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

A Kind of new parallax algorithm based on symmetric continuous optimization

引用

Journal of Physics: Conference Series 2021年第3期1748卷

作者： Kaiwen Jiang Degan Zhang Haixia Xu Key Laboratory of Computer Vision and System (Tianjin University of Technology) Ministry of Education 300384 China Tianjin Key Lab of Intelligent Computing & Novel software Technology Tianjin University of Technology Tianjin China School of Electronic and Information Engineering Tianjin Vocational Institute Tianjin 300410 China

With the fast development of 3D imaginations becomes more and more fascination, multi-view stereo based 3D reconstruction is a significant technique for those application. To facilitate the subsequent processing of 3D reconstruction and reduce the possibility of other algorithms falling into local optimal solutions, attempting to get better and faster performance, a new parallax calculation based on symmetric continuous optimization is proposed in this paper. The algorithm is proposed here will be tested respectively in the same data, in order to certificate the algorithm applied is better than traditional minimize El algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The improvement of wireless LAN security authentication mechanism based on Kerberos

The improvement of wireless LAN security authentication mech...

引用

2018 International Conference on Electronics technology, ICET 2018

作者： Ma, Yi Ning, Hongyun Tianjin Key Laboratory of Intelligence Computing and New Software Technology Tianjin University of Technology Tianjin China

ISBN: (纸本)9781538657522

In order to solve the problem of vulnerable password guessing attacks caused by dictionary attacks, replay attacks in the authentication process, and man-in-The-middle attacks in the existing wireless local area network in terms of security authentication, we make some improvements to the 802.1X / EAP authentication protocol based on the study of the current IEEE802.11i security protocol with high security. After introducing the idea of Kerberos protocol authentication and applying the idea in the authentication process of 802.1X / EAP, a new protocol of Kerberos extensible authentication protocol (KEAP) is proposed. Firstly, the protocol introduces an asymmetric key encryption method, uses public key encryption during data transmission, and the receiver uses the corresponding private key for decryption. With unidirectional characteristics and high security, the encryption can avoid password guessing attacks caused by dictionary attacks as much as possible. Secondly, aiming at the problem that the request message sent from the client to the authentication server is vulnerable to replay attacks, the protocol uses a combination of the message sequence number and the random number, and the message serial number is added to the request message sent from the client to the authentication server. And establish a list database for storing message serial number and random number in the authentication server. After receiving a transfer message, the serial number and the random number are extracted and compared with the values in the list database to distinguish whether it is a retransmission message. Finally, the protocol introduces a keychain mechanism and uses an irreversible Hash function to encrypt the final authentication result, thereby effectively solving the man-in-The-middle attack by the pretender. The experiment uses the OPNET 14.5 simulation platform to model the KEAP protocol and simulate simulation attacks, and compares it with the current more common EAP-TLS au

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Deep Pyramid Convolutional Neural Network Integrated with Self-attention Mechanism and Highway Network for Text Classification

引用

Journal of Physics: Conference Series 2020年第1期1642卷

作者： Xuewei Li Hongyun Ning Tianjin Key Laboratory of Intelligence Computing & New Software Technology Tianjin University of Technology Tianjin300384 China

Text classification is one of the basic tasks of natural language processing. In recent years, deep learning has been widely used in text classification tasks. The representative one is the convolutional neural network. DPCNN is a deep convolutional neural network text classification model that can obtain long-distance text information, but it focuses on the extraction of global features and ignores the extraction of local features of the text. Some local feature information is very important and plays an important role in text classification tasks. Therefore, in this paper, a self-attention mechanism is introduced to extract local features of text based on the lack of extracting local features of DPCNN. In addition, although the deep convolutional neural network can extract deeper features, it is easy to cause the problem of gradient disappearance during training. Therefore, the highway network is introduced to prevent the problem of gradient disappearance caused by network training and improve the performance of the model. Experimental results show that the proposed model is better than a single DPCNN model, which further improves the accuracy of text classification tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

newBalance: Efficient Data Space Management and Algorithmic Optimization for Large-Scale Storage Systems

引用

Chinese Journal of Electronics 2017年第3期26卷 493-501页

作者： XU Guangping LIN Sheng SHI Kai ZHANG Hua Tianjin Key Laboratory of Intelligence Computing and New Software Technology Computer and Communication Engineering School Tianjin University of Technology

Fragmentation usually occurs when data space of original storage nodes has to be reallocated to new added storage nodes during the scale-out evolution of the large-scale storage *** greatly influences its performance and becomes a challenge to manage the whole *** present an efficient space management framework,called new Balance,to reduce fragmentation with the minimum data movement while keeping the storage system load *** space management framework has two phases including the collection phase and the allocation *** the collection phase,we propose a novel algorithm,called the greedy bi-direction collector,which collects enough space for the new storage *** the allocation phase,we formally represent it as a variant of the bin packing problem and then utilize some bin packing heuristics including the first fitting and the best fitting to allocate collected intervals to new added storage *** experimental results show that the amount of intervals can be reduced by 20%;5%and our algorithmic optimization improves the data lookup performance by at least 10%and the scale-out performance by 2X;X.

关键词： Large-scale storage systems Space management Algorithmic optimization Performance evaluation

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised classification method of SAR images using spectral clustering in contourlet domain

引用

Journal of Physics: Conference Series 2020年第4期1486卷

作者： Jiang, Kaiwen Zhang, Degan Xu, Haixia Key Laboratory of Computer Vision and System (Tianjin University of Technology) Ministry of Education 300384 China Key Laboratory of Computer Vision and System (Tianjin University of Technology) Ministry of Education 300384 China Tianjin Key Lab of Intelligent Computing & Novel software Technology Tianjin University of Technology Tianjin China

A new based on Semi-supervised classification theory for SAR images in contourlet domain is proposed, in this paper. Attempting to get better and faster performance, the PSO algorithm (Particle swarm optimization algorithm) and contourlet domain is proposed to instead of traditional k-means algorithm. PSO is used to find the global optimum by performing a global search in the whole solution space. And then, contourlet is applied in front of construct the similarity matrix to extract more effective eigenvalues. In section five, the proposed algorithm got better classification results than the traditional k-means algorithm which is proved by experimental results show that in terms of running time, classification accuracy and Kappa coefficient.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ecg Based Identification Under Different Physical Status

Ecg Based Identification Under Different Physical Status

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Gang Zheng Hong-Yuan Zhang Shen-Zhen Ji Tianjin Key Laboratory of Intelligent Computing and Software New Technique Tianjin University of Technology Tianjin China

ISBN: (纸本)9781538652152

Relatively stable feature of Electrocardiogram (ECG) waveform in ECG based identification authentication is an important issue to be discovered in biometric. It can benefit the biomaterial which is weak and easy to be affected when it is collected under different physical status. From the paper, it was found that P waveform was stable when several ECG waveforms were overlapped under Matthew Effect rule, therefore P waveform period could be more dependent than other part of ECG waveform in identity. And this result was used in the following experiments. The experiments in the paper were designed to study whether ECG based identification was affected by the status of ECG signal collecting. The status contained four status pairs, that was, emotion status (calm, high pressure), eating (starve, satition), sleeping (full, lack) and health (healthy, tired/cold). From the result, Negative emotion and lacking sleeping reduced TPR slightly (about 2-3%), but there was no effect on both eating and health pair status. However, if we used ECG data mixed with all four statuses, the overall True Positive Rate (TPR) of identification reached around 85%.

关键词： Electrocardiography Authentication Training Heart beat Feature extraction Shape Frequency-domain analysis

来源：评论

学校读者我要写书评

暂无评论

On-line Detection and Analysis of Alloy Steel Elements Based on the LIBS technology and Random Forest Regression

On-line Detection and Analysis of Alloy Steel Elements Based...

引用

第37届中国控制会议

作者： Yang Jixuan Zhang Qin Yang Peng Sun Junqing Tianjin Key Laboratory of Intelligent Computing and Novel Software Technology Tianjin University of Technology

The Laser Induced Breakdown Spectroscopy（LIBS） technology can be used to detect the elements in the alloy steel in real time. Quantitative analysis method of the traditional LIBS technology mainly has the calibration method and calibration free method, but there are two shortcomings: low prediction accuracy and over fitting. Random Forest Regression（RFR） algorithm can be used for classification and regression, can effectively avoid "overfitting" phenomenon. Therefore, in this paper, we combine the random forest regression algorithm with laser induced breakdown spectroscopy applied to the detection of the concentration of alloy steel elements in the metallurgy industry. At the same time, compared with partial least squares method based on the LIBS, the results show that the random forest algorithm combined with the LIBS technology has the higher prediction accuracy, lower root mean square error and better robustness.

关键词： LIBS quantitative analysis RFR PLS root mean square error

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：