检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Wu, Jiasong Zhang, Jing Wu, Fuzhi Kong, Youyong Yang, Guanyu Senhadji, Lotfi Shu, Huazhong Laboratory of Image Science and Technology Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing210096 China Université de Rennes INSERM LTSI-UMR 1099 RennesF-35042 France Université de Rennes INSERM Southeast University Rennes France

Deep learning is a hot research topic in the field of machine learning methods and applications. Generative Adversarial networks (GANs) and Variational Auto-Encoders (VAEs) provide impressive image generations from Gaussian white noise, but both of them are difficult to train since they need to train the generator (or encoder) and the discriminator (or decoder) simultaneously, which is easy to cause unstable training. In order to solve or alleviate the synchronous training difficult problems of GANs and VAEs, recently, researchers propose Generative Scattering networks (GSNs), which use wavelet scattering networks (ScatNets) as the encoder to obtain the features (or ScatNet embeddings) and convolutional neural networks (CNNs) as the decoder to generate the image. The advantage of GSNs is the parameters of ScatNets are not needed to learn, and the disadvantage of GSNs is that the expression ability of ScatNets is slightly weaker than CNNs and the dimensional reduction method of Principal Component Analysis (PCA) is easy to lead overfitting in the training of GSNs, and therefore affect the generated quality in the testing process. In order to further improve the quality of generated images while keep the advantages of GSNs, this paper proposes Generative Fractional Scattering networks (GFRSNs), which use more expressive fractional wavelet scattering networks (FrScatNets) instead of ScatNets as the encoder to obtain the features (or FrScatNet embeddings) and use the similar CNNs of GSNs as the decoder to generate the image. Additionally, this paper develops a new dimensional reduction method named Feature-Map Fusion (FMF) instead of PCA for better keeping the information of FrScatNets and the effect of image fusion on the quality of image generation is also discussed. The experimental results on CIFAR-10 dataset and CelebA dataset show that the proposed GFRSNs can obtain better generated images than the original GSNs on the testing dataset. Copyright © 2020, The Author

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Encrypted traffic identification based on N-gram entropy and cumulative sum test 2018

Encrypted traffic identification based on N-gram entropy and...

引用

13th International Conference on Future Internet Technologies, CFI 2018

作者： Cheng, Guang Hu, Ying School of Cyber Science and Engineering Southeast University China Key Laboratory of Computer Network and Information Integration Ministry of Education China

ISBN: (纸本)9781450364669

Since existing methods using entropy are less effective in characterizing encrypted traffic, this paper proposes an encrypted traffic identification method based on n-gram entropy and cumulative sum. This method analyzes the entropy characteristics of n-gram entropy for text, picture, compressed file, and encrypted traffic in the network. Furthermore, an analysis of cumulative sum is performed to better distinguish compressed file traffic and encrypted traffic. The experiments show that our propsed method reaches high accuracy for encrypted traffic identification and performs well in distinguishing compressed file traffic and encrypted traffic. © 2018 Association for Computing Machinery.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Mining hypernym-hyponym relations from social tags via tag embedding 5th

Mining hypernym-hyponym relations from social tags via tag e...

引用

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

作者： Zhang, Mengyi Wu, Tianxing Ji, Qiu Qi, Guilin Sun, Zhixin School of Computer Science and Engineering Southeast University Nanjing China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing China School of Computer Science and Engineering Nanyang Technological University Singapore Singapore School of Modern Posts and Institute of Modern Posts Nanjing University of Posts and Telecommunications Nanjing China YuanTong Express Co. LTD. Shanghai China

ISBN: (纸本)9783030242701

With the rapid development of Internet of Thing, mobile Internet, cloud computing and other technologies, network data increases dramatically and Folksonomy plays an important role in web systems. How to obtain valuable knowledge, especially hypernym-hyponym relations, becomes a popular research topic in the field of artificial intelligence. For Folksonomy, hypernym-hyponym relation identification aims to recognize the"is-a" relation between two social tags. Most existing works about identifying hypernym-hyponym relations are based on statistical and heuristic approaches, but their performance still needs to be improved. In this paper, we propose a novel supervised learning approach to identify hypernym-hyponym relations from social tags using tag embeddings. First, we use a neural network model to learn tag embeddings. This model relies on not only the hypernym and hyponym tags, but also the contextual information between them. We then apply such embeddings as features to identify hypernym-hyponym relations using a supervised learning method. Our experimental results demonstrate that the proposed approach significantly outperforms other state-of-the-art approaches over a labeled dataset. The accuracy and F1-score of our approach achieve 0.91 and 0.86 respectively. © Springer Nature Switzerland AG 2019.

关键词： Heuristic methods

来源：评论

学校读者我要写书评

暂无评论

Optimal Defense Policies for Cost-Constrained Cyber Physical System

Optimal Defense Policies for Cost-Constrained Cyber Physical...

引用

International Conference on Advanced Cloud and Big Data (CBD)

作者： Yong Cheng Yonggang Yang Qian Guo Bingfeng Xu Gaofeng He State Grid Shanxi Eelectric Power Company Xian China Global Energy Interconnection Research Institute co. Ltd Nanjing China College of Information Science and Technology Nanjing Forestry University Nanjing China College of Internet of Things Nanjing University of Posts and Telecommunications Nanjing China Ministry of Education Key Laboratory of Computer Network and Information Integration (Southeast University)

ISBN: (纸本)9781665423137;9781665446273

Cyber-Physical System (CPS) usually consists of vast amounts of heterogeneous nodes, and several types of nodes may not be upgraded or installed with new programs during their life spans. How to choose appropriate nodes to deploy security defense mechanisms to ensure system security is a vital and challenging issue. This paper considers the selection of optimal defense policies under a given cost constraint. First, we use the Attack Defense Tree (ADTree) to model attacks and countermeasures for CPS. Then, each defense node in ADTree is given a defense cost attribute and a defense success probability attribute as security evaluation indicators. After that, a probability tree model is proposed for evaluating the final defense success probability. Finally, based on the Minimum Cut Set (MCS) theory, the nodes that can maximize the defense probability with acceptable cost are selected as the optimal defense nodes. An open-source calculation tool is implemented, and the evaluation is performed through a denial of service attack against CPS. The results show that the proposed method is accurate and efficient, which can guide the formulation of CPS security protection.

关键词： Tools Cyber-physical systems Big Data Bayes methods Security Iterative methods Task analysis

来源：评论

学校读者我要写书评

暂无评论

Building an efficient intrusion detection system based on feature selection and ensemble classifier

arXiv

引用

arXiv 2019年

作者： Zhou, Yuyang Cheng, Guang Jiang, Shanqing Dai, Mian School of Cyber Science and Engineering Southeast University Nanjing China Key Laboratory of Computer Network and Information Integration Ministry of Education Nanjing China Jiangsu Provincial Key Laboratory of Computer Network Technology Southeast University Nanjing China National Key Laboratory of Science and Technology on Information System Security Beijing China

Intrusion detection system (IDS) is one of extensively used techniques in a network topology to safeguard the integrity and availability of sensitive assets in the protected systems. Although many supervised and unsupervised learning approaches from the field of machine learning have been used to increase the efficacy of IDSs, it is still a problem for existing intrusion detection algorithms to achieve good performance. First, lots of redundant and irrelevant data in high-dimensional datasets interfere with the classification process of an IDS. Second, an individual classifier may not perform well in the detection of each type of attacks. Third, many models are built for stale datasets, making them less adaptable for novel attacks. Thus, we propose a new intrusion detection framework in this paper, and this framework is based on the feature selection and ensemble learning techniques. In the first step, a heuristic algorithm called CFS-BA is proposed for dimensionality reduction, which selects the optimal subset based on the correlation between features. Then, we introduce an ensemble approach that combines C4.5, Random Forest (RF), and Forest by Penalizing Attributes (Forest PA) algorithms. Finally, voting technique is used to combine the probability distributions of the base learners for attack recognition. The experimental results, using NSL-KDD, AWID, and CIC-IDS2017 datasets, reveal that the proposed CFS-BA-Ensemble method is able to exhibit better performance than other related and state of the art approaches under several metrics. Copyright © 2019, The Authors. All rights reserved.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Neural storyline extraction model for storyline generation from news articles

Neural storyline extraction model for storyline generation f...

引用

2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2018

作者： Zhou, Deyu Guo, Linsen He, Yulan School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration Southeast University China School of Engineering and Applied Science Aston University United Kingdom

ISBN: (纸本)9781948087278

Storyline generation aims to extract events described on news articles under a certain topic and reveal how those events evolve over time. Most existing approaches first train supervised models to extract events from news articles published in different time periods and then link relevant events into coherent stories. They are domain dependent and cannot deal with unseen event types. To tackle this problem, approaches based on probabilistic graphic models jointly model the generations of events and storylines without annotated data. However, the parameter inference procedure is too complex and models often require long time to converge. In this paper, we propose a novel neural network based approach to extract structured representations and evolution patterns of storylines without using annotated data. In this model, title and main body of a news article are assumed to share the similar storyline distribution. Moreover, similar documents described in neighboring time periods are assumed to share similar storyline distributions. Based on these assumptions, structured representations and evolution patterns of storylines can be extracted. The proposed model has been evaluated on three news corpora and the experimental results show that it outperforms state-of-The-Art approaches accuracy and efficiency. © 2018 The Association for Computational Linguistics.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Generalization Based Privacy-Preserving Provenance Publishing 15th

Generalization Based Privacy-Preserving Provenance Publishin...

引用

15th Web information Systems and Applications Conference, WISA 2018

作者： Wu, Jian Ni, Weiwei Zhang, Sen Department of Computer Science and Engineering Southeast University Nanjing211189 China Key Laboratory of Computer Network and Information Integration in Southeast University Ministry of Education Nanjing211189 China

ISBN: (纸本)9783030029333

With thriving of data sharing, demands of data provenance publishing become increasingly urgent. Data provenance describes about how data is generated and evolves with time. Data provenance has many applications, in-cluding evaluation of data quality, audit trail, replication recipes, data citation, etc. Some in-out mapping relations and related intermediate parameters in data provenance may be private. How to protect the privacy in the data provenance publishing attracts increasing attention from researchers in recent years. Existing solutions rely primarily on Γ-privacy model, hiding certain properties to solve the module’s privacy-preserving problem. However, the Γ-privacy model has the following disadvantages: (1) The attribute domains are limited. (2) It’s difficult to set consistent Γ value for the workflow. (3) The attribute selection strategy is unreasonable. Concerning these problems, a novel privacy-preserving provenance model is devised to balance the tradeoff between privacy-preserving and utility of data provenance. The devised model applies the generalization and introduces the generalized level. Furthermore, an effective privacy-preserving provenance publishing method based on generalization is proposed to achieve the privacy security in the data provenance publishing. Finally, theoretical analysis and experimental results testifies the effectiveness of our solution. © 2018, Springer Nature Switzerland AG.

关键词： Quality control

来源：评论

学校读者我要写书评

暂无评论

COSINE:a software development model integrating collective intelligence, service and ecosystem

COSINE:a software development model integrating collective i...

引用

IEEE World Congress on Services (SERVICES)

作者： Hongbing Wang Tianjing Hong Jian Cao Haijun Zhang Changhai Nie Bo Cheng Yangfan He Li Kuang Dunwei Gong Wuhui Chen Yuliang Shi Deyi Huang School of Computer Science and Engineering and Key Laboratory of Computer Network and Information Integration Southeast University Nanjing China Shanghai Jiaotong University Shanghai China College of Computer Science and Technology Harbin Institute of Technology Shenzhen China Nanjing University Nanjing China Wuhan University Wuhan China Research Institute of Network Technology Beijing University of Posts and Telecommunications Beijing China Software College Central South University Changsha China College of information and control engineering China University of mining and technology Xuzhou China College of data science and computer science Zhongshan University Guangzhou China Shandong University Jinan China Landry Company Shenzhen China

ISBN: (数字)9781728182032

ISBN: (纸本)9781728182049

With the development of the internet technology, a large amount of softwares have emerged to meet users' increasing needs. At the mean time, software systems have been faced with a problem that they must adapt to the dynamic network environment. It is obvious that a variety of software development models have been proposed in the past few decades. However, the majority of these methods are gradually unadaptable to new circumstances. In this paper, we proposed a new software development model integrating collective intelligence, service and ecosystem. On the one hand, we have introduced the model in detail. On the other hand, We took a practical example to demonstrate the effectiveness of the proposed model.

关键词： Adaptation models Biological system modeling Ecosystems Collective intelligence Software systems

来源：评论

学校读者我要写书评

暂无评论

Relevant emotion ranking from text constrained with emotion relationships

Relevant emotion ranking from text constrained with emotion ...

引用

2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2018

作者： Zhou, Deyu Yang, Yang He, Yulan School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University China School of Engineering and Applied Science Aston University United Kingdom

ISBN: (纸本)9781948087278

Text might contain or invoke multiple emotions with varying intensities. As such, emotion detection, to predict multiple emotions associated with a given text, can be cast into a multi-label classification problem. We would like to go one step further so that a ranked list of relevant emotions are generated where top ranked emotions are more intensely associated with text compared to lower ranked emotions, whereas the rankings of irrelevant emotions are not important. A novel framework of relevant emotion ranking is proposed to tackle the problem. In the framework, the objective loss function is designed elaborately so that both emotion prediction and rankings of only relevant emotions can be achieved. Moreover, we observe that some emotions co-occur more often while other emotions rarely coexist. Such information is incorporated into the framework as constraints to improve the accuracy of emotion detection. Experimental results on two real-world corpora show that the proposed framework can effectively deal with emotion detection and performs remarkably better than the state-of-The-Art emotion detection approaches and multi-label learning methods. © 2018 The Association for Computational Linguistics.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Analysis and Improvement of an Efficient Controlled Quantum Secure Direct Communication and Authentication Protocol

引用

computers, Materials & Continua 2018年第12期57卷 621-633页

作者： Jifeng Zhong Zhihao Liu Juan Xu Navigation College Jimei UniversityXiamen361021China School of Computer Science and Engineering Southeast UniversityNanjing211189China Key Laboratory of Computer Network and Information Integration Southeast UniversityMinistry of EducationNanjing211189China Institute for Quantum Computing University of WaterlooWaterlooONCanadaN2L3G1

The controlled quantum secure direct communication(CQSDC)with authentication protocol based on four particle cluster states via quantum one-time pad and local unitary operations is *** is found that there are some serious security issues in this *** eavesdropper(Eve)can eavesdrop on some information of the identity strings of the receiver and the controller without being detected by the selective-CNOT-operation(SCNO)*** the same attack,Eve can also steal some information of the secret message that the sender *** addition,the receiver can take the same kind of attack to eavesdrop on some information of the secret message out of the control of the *** means that the requirements of CQSDC are not *** last,we improve the original CQSDC protocol to a secure one.

关键词： Quantum cryptography controlled quantum secure direct communication selective-CNOT-operation attack

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：