Resource Type

Journal Article 415

Conference Videos 56

Conference Information 44

Conference Topics 2

Year

2023 39

2022 85

2021 47

2020 62

2019 49

2018 26

2017 30

2016 20

2015 7

2014 6

2013 5

2012 6

2011 8

2010 7

2009 6

2008 6

2007 10

2006 13

2005 21

2004 18

open ︾

Keywords

informatization 7

information technology 6

pattern recognition 6

CAPP 3

Cyber-physical system 3

Geographic information system 3

Machine learning 3

Artificial intelligence 2

Blockchain 2

Cloud computing 2

Cyber-physical systems 2

Database schemata 2

Information schema constructs 2

Intelligent manufacturing 2

Internet Plus 2

Internet of Things 2

Multi-objective optimization 2

Random forest 2

Reinforcement learning 2

open ︾

Search scope:

排序： Display mode:

一种易用的实体识别消歧系统评测框架 Article

辉 陈,宝刚 魏,一鸣 李,Yong-huai LIU,文浩 朱

Frontiers of Information Technology & Electronic Engineering 2017, Volume 18, Issue 2,   Pages 195-205 doi: 10.1631/FITEE.1500473

Abstract: 实体识别消歧是知识库扩充和信息抽取的重要技术之一。近些年该领域诞生了很多研究成果，提出了许多实体识别消歧系统。但由于缺乏对这些系统的完善评测对比，该领域依然处于良莠淆杂的状态。因此很有必要设计一个评测框架对各个系统进行统一评测。本文提出一个实体识别消歧系统的统一评测框架，用于公平地比较各个实体识别消歧系统的效果。该框架代码开源，可以采用新的系统、数据集、评测机制扩展。通过该框架评测实体系统，可以分析得到系统各个模块的优劣之处。本文分析对比了几个公开的实体识别消歧系统，并总结出了一些有用的结论。

Keywords： 实体识别消歧；评测框架；信息抽取    

HTML PDF Collect

Disambiguating named entitieswith deep supervised learning via crowd labels Article

Le-kui ZHOU,Si-liang TANG,Jun XIAO,Fei WU,Yue-ting ZHUANG

Frontiers of Information Technology & Electronic Engineering 2017, Volume 18, Issue 1,   Pages 97-106 doi: 10.1631/FITEE.1601835

Abstract: Named entity disambiguation (NED) is the task of linking mentions of ambiguous entities to their referenced entities in a knowledge base such as Wikipedia. We propose an approach to effectively disentangle the discriminative features in the manner of collaborative utilization of collective wisdom (via human-labeled crowd labels) and deep learning (via human-generated data) for the NED task. In particular, we devise a crowd model to elicit the underlying features (crowd features) from crowd labels that indicate a matching candidate for each mention, and then use the crowd features to fine-tune a dynamic convolutional neural network (DCNN). The learned DCNN is employed to obtain deep crowd features to enhance traditional hand-crafted features for the NED task. The proposed method substantially benefits from the utilization of crowd knowledge (via crowd labels) into a generic deep learning for the NED task. Experimental analysis demonstrates that the proposed approach is superior to the traditional hand-crafted features when enough crowd labels are gathered.

Keywords： Named entity disambiguation     Crowdsourcing     Deep learning    

HTML PDF Collect

A review on cyber security named entity recognition Review Article

Chen Gao, Xuan Zhang, Mengting Han, Hui Liu,zhxuan@ynu.edu.cn

Frontiers of Information Technology & Electronic Engineering 2021, Volume 22, Issue 9,   Pages 1153-1168 doi: 10.1631/FITEE.2000286

Abstract: With the rapid development of Internet technology and the advent of the era of big data, more and more texts are provided on the Internet. These texts include not only security concepts, incidents, tools, guidelines, and policies, but also risk management approaches, best practices, assurances, technologies, and more. Through the integration of large-scale, heterogeneous, unstructured information, the identification and classification of entities can help handle issues. Due to the complexity and diversity of texts in the domain, it is difficult to identify security entities in the domain using the traditional methods. This paper describes various approaches and techniques for NER in this domain, including the rule-based approach, dictionary-based approach, and based approach, and discusses the problems faced by NER research in this domain, such as conjunction and disjunction, non-standardized naming convention, abbreviation, and massive nesting. Three future directions of NER in are proposed: (1) application of unsupervised or semi-supervised technology; (2) development of a more comprehensive ontology; (3) development of a more comprehensive model.

Keywords： 命名实体识别（NER）；信息抽取；网络空间安全；机器学习；深度学习    

HTML PDF Collect

A network security entity recognition method based on feature template and CNN-BiLSTM-CRF Research Papers

Ya QIN, Guo-wei SHEN, Wen-bo ZHAO, Yan-ping CHEN, Miao YU, Xin JIN

Frontiers of Information Technology & Electronic Engineering 2019, Volume 20, Issue 6,   Pages 872-884 doi: 10.1631/FITEE.1800520

Abstract:

By network security threat intelligence analysis based on a security knowledge graph (SKG), multi-source threat intelligence data can be analyzed in a fine-grained manner. This has received extensive attention. It is difficult for traditional named entity recognition methods to identify mixed security entities in Chinese and English in the field of network security, and there are difficulties in accurately identifying network security entities because of insufficient features extracted. In this paper, we propose a novel FT-CNN-BiLSTM-CRF security entity recognition method based on a neural network CNN-BiLSTM-CRF model combined with a feature template (FT). The feature template is used to extract local context features, and a neural network model is used to automatically extract character features and text global features. Experimental results showed that our method can achieve an F-score of 86% on a large-scale network security dataset and outperforms other methods.

Keywords： Network security entity     Security knowledge graph (SKG)     Entity recognition     Feature template     Neural network    

HTML PDF Collect

A Face Recognition Based on Fusion Features Extraction From Two Kinds of Projection

Zhang Shengliang,Xu Yong,Yang Jian,Yang Jingyu

Strategic Study of CAE 2006, Volume 8, Issue 8,   Pages 50-55

Abstract:

A novel face recognition algorithm based on two kinds of projection is presented in this paper. First, the two dimension principal component analysis (2DPCA) is used to extract one group of features, denoted by α. Second, the fisher linear discriminant analysis (LDA) , or fisherfaces, is used for extracting another group of features, denoted by β.After being standardized, the two kinds of features are combined together in the form of the complex vector α+iβ. Then the fusion features in the complex feature space is extracted by using complex PCA (CPCA). The proposed algorithm is evaluated by using the FERET face database at three different resolutions. The experimental results indicate that the proposed method can achieve about 10% higher recognition accurate rate than 2DPCA and LDA, while only using 28 features for each sample.

Keywords： feature fusion     linear discriminant analysis (LDA)     feature extraction     face recognition    

HTML PDF Collect

Learning to select pseudo labels: a semi-supervised method for named entity recognition Research Articles

Zhen-zhen Li, Da-wei Feng, Dong-sheng Li, Xi-cheng Lu,lizhenzhen14@nudt.edu.cn,davyfeng.c@gmail.com,dsli@nudt.edu.cn,xclu@nudt.edu.cn

Frontiers of Information Technology & Electronic Engineering 2020, Volume 21, Issue 6,   Pages 809-962 doi: 10.1631/FITEE.1800743

Abstract: models have achieved state-of-the-art performance in (NER); the good performance, however, relies heavily on substantial amounts of labeled data. In some specific areas such as medical, financial, and military domains, labeled data is very scarce, while is readily available. Previous studies have used to enrich word representations, but a large amount of entity information in is neglected, which may be beneficial to the NER task. In this study, we propose a for NER tasks, which learns to create high-quality labeled data by applying a pre-trained module to filter out erroneous pseudo labels. Pseudo labels are automatically generated for and used as if they were true labels. Our semi-supervised framework includes three steps: constructing an optimal single neural model for a specific NER task, learning a module that evaluates pseudo labels, and creating new labeled data and improving the NER model iteratively. Experimental results on two English NER tasks and one Chinese clinical NER task demonstrate that our method further improves the performance of the best single neural model. Even when we use only pre-trained static word embeddings and do not rely on any external knowledge, our method achieves comparable performance to those state-of-the-art models on the CoNLL-2003 and OntoNotes 5.0 English NER tasks.

Keywords： 命名实体识别；无标注数据；深度学习；半监督学习方法    

HTML PDF Collect

A Study on the Essence of Optimal Statistical Uncorrelated Discriminant Vectors

Wu Xiaojun,Yang Jingyu,Wang Shitong,Liu Tongming,Josef Kittler

Strategic Study of CAE 2004, Volume 6, Issue 2,   Pages 44-47

Abstract:

A study has been made on the essence of optimal set of uncorrelated discriminant vectors in this paper. A whitening transform has been constructed on the basis of the eigen decomposition of population scatter matrix, which makes the population scatter matrix an identity matrix in the transformed sample space. Thus, the optimal discriminant vectors solved by conventional LDA methods are statistical uncorrelated. The research indicates that the essence of the statistical uncorrelated discriminant transform is the whitening transform plus conventional linear discriminant transform. The distinguished characteristic of the proposed method is that the obtained optimal discriminant vectors are orthogonal and statistical uncorrelated. The proposed method suits for all the problems of algebraic feature extraction. The numerical experiments on facial database of ORL show the effectiveness of the proposed method.

Keywords： pattern recognition     feature extraction     disciminant analysis     generalized optimal set of discriminant vectors     face recognition    

HTML PDF Collect

Both-Branch Fuzzy Decision and Problems on Decision Discernment

Shi Kaiquan,Li Qiqiang

Strategic Study of CAE 2001, Volume 3, Issue 1,   Pages 71-77

Abstract:

This paper proposes the concept of both-branch fuzzy decision on X, which contains neutral universe(X*≠{x}), and the optimal decision analysis model. In addition, the paper proposes decision judgement theorem, decision discernment theorem, decision surplusage-discarding theorem and hole-digging principle on decision factors universe X. Both-branch fuzzy decision has such characteristics as decision two-direction dependence character, decision piling-synthesis character, decision branch-separating character, decision branch-degeneration character, and decision non-fault character. The research results have found application.

Keywords： both-branch fuzzy decision     decision model     judgement theorem     discernment theorem     surplusage-discarding theorem     hole-digging principle    

HTML PDF Collect

High-Temperature Target Recognition Based on Spectral Radiation Information

Fan Xueliang,Cheng Xiaofang,Xu Jun

Strategic Study of CAE 2004, Volume 6, Issue 6,   Pages 57-62

Abstract:

Based on the principles of optics and radiometry, the imaging mathematical model is established and the factors of the contrast (signal-noise-ratio) of high-temperature target and the scenery are given. By analyzing not only the differences in spectral properties between objects in the scene, but also the CCD spectral response theoretically, a new method of enhancement of contrast is given. By optimizing the initial image capture stage, using liquid crystal light valve to make a simple modification of the imaging system, the goal of high object recognition is achieved. The experimental results agree with the theoretical predicts.

Keywords： video image     object recognition     radiation information     liquid crystal light valve    

HTML PDF Collect

DDUC: an erasure-coded system with decoupled data updating and coding Research Article

Xiang LI, Yibing LI, Chunrui TANG, Yingsong LI,chunruitang@126.com

Frontiers of Information Technology & Electronic Engineering 2023, Volume 24, Issue 5,   Pages 742-758 doi: 10.1631/FITEE.2200253

Abstract: To improve the accuracy of modulated signal recognition in variable environments and reduce the impact of factors such as lack of prior knowledge on recognition results, researchers have gradually adopted deep learning techniques to replace traditional modulated signal processing techniques. To address the problem of low recognition accuracy of the modulated signal at low signal-to-noise ratios, we have designed a novel network of multi-scale analysis with deep threshold noise elimination to recognize the actually collected modulated signals under a symmetric cross-entropy function of label smoothing. The network consists of a denoising encoder with deep adaptive threshold learning and a decoder with . The two modules are skip-connected to work together to improve the robustness of the overall network. Experimental results show that this method has better recognition accuracy at low signal-to-noise ratios than previous methods. The network demonstrates a flexible self-learning capability for different noise thresholds and the effectiveness of the designed feature fusion module in multi-scale feature acquisition for various modulation types.

Keywords： Signal noise elimination     Deep adaptive threshold learning network     Multi-scale feature fusion     Modulation recognition    

HTML PDF Collect

A Framework of Knowledge Theory： Toward a Unified Theory of Information, Knowledge and Intelligence

Zhong Yixin

Strategic Study of CAE 2000, Volume 2, Issue 9,   Pages 50-64

Abstract:

Knowledge has been very important wealth to the mankind but there has not a knowledge theory existed yet till the present time. An attempt is thus made in the paper to present a framework of knowledge theory that includes two parts: fundamentals and the main body of knowledge theory. The first part is to deal with a series of basic issues such as the related concepts and definitions, the methods of representation, the measurements, the reasoning and decision rules. The second part is to explore the mechanism of knowledge formation based on information processing and the mechanism of intelligence formation based on the activation of knowledge. It is believed that the establishment of the knowledge theory will lay a solid foundation to the unified theory of information, knowledge, and intelligence and will greatly facilitate the effective utilization of information and knowledge, leading to the growth of the research in the field of intelligent machines.

Keywords： knowledge     amount of knowledge     knowledge formation     knowledge activation     unified theory of information-knowledge-intelligence    

HTML PDF Collect

Algorithm Design for Improving Feature Extraction Efficiency Based on KPCA

Xu Yong,Yangjingyu,Lu Jianfeng

Strategic Study of CAE 2005, Volume 7, Issue 10,   Pages 38-42

Abstract:

KPCA (kernel PCA) is derived from PCA. It can extract nonlinear feature components of samples. However, feature extraction for one sample requires that kernel functions between training samples and the sample be calculated in advance. So, the size of training sample set affects the efficiency of feature extraction. It is supposed that in feature space the eigenvectors may be linearly expressed by a part of training samples, called nodes. According to the supposition, an improved KPCA (IKPCA) algorithm is developed. IKPCA extracts feature components of one sample efficiently, only based on kernel functions between nodes and the sample. Experimental results show that IKPCA is very close to KPCA in performance, while with higher efficiency.

Keywords： KPCA(Kernel PCA)     IKPCA(Improved KPCA)     feature extraction     feature space    

HTML PDF Collect

Improving entity linking with two adaptive features Research Article

Hongbin ZHANG, Quan CHEN, Weiwen ZHANG

Frontiers of Information Technology & Electronic Engineering 2022, Volume 23, Issue 11,   Pages 1620-1630 doi: 10.1631/FITEE.2100495

Abstract:

(EL) is a fundamental task in natural language processing. Based on neural networks, existing systems pay more attention to the construction of the , but ignore latent semantic information in the and the acquisition of effective information. In this paper, we propose two , in which the first adaptive feature enables the local and s to capture latent information, and the second adaptive feature describes effective information for embeddings. These can work together naturally to handle some uncertain information for EL. Experimental results demonstrate that our EL system achieves the best performance on the AIDA-B and MSNBC datasets, and the best average performance on out-domain datasets. These results indicate that the proposed , which are based on their own diverse contexts, can capture information that is conducive for EL.

Keywords： Entity linking     Local model     Global model     Adaptive features     Entity type    

HTML PDF Collect

Information schema constructs for defining warehouse databases of genotypes and phenotypes of system manifestation features Article

Shahab POURTALEBI,Imre HORVÁTH

Frontiers of Information Technology & Electronic Engineering 2016, Volume 17, Issue 9,   Pages 862-884 doi: 10.1631/FITEE.1600997

Abstract: Our long-term objective is to develop a software toolbox for pre-embodiment design of complex and heterogeneous systems, such as cyber-physical systems. The novelty of this toolbox is that it uses system manifestation features (SMFs) for transdisciplinary modeling of these systems. The main challenges of implementation of the toolbox are functional design- and language-independent computational realization of the warehouses, and systematic development and management of the various evolving implements of SMFs (genotypes, phenotypes, and instances). Therefore, an information schema construct (ISC) based approach is proposed to create the schemata of the associated warehouse databases and the above-mentioned SMF implements. ISCs logically arrange the data contents of SMFs in a set of relational tables of varying semantics. In this article we present the ISCs necessary for creation of genotypes and phenotypes. They increase the efficiency of the database development process and make the data relationships transparent. Our follow-up research focuses on the elaboration of the SMF instances based system modeling methodology.

Keywords： Cyber-physical systems     Software toolbox     Pre-embodiment design     System manifestation features (SMFs)     Warehouses     Database schemata     SMF genotypes     SMF phenotypes     SMF instances     Information schema constructs    

HTML PDF Collect

Research on evaluation of extinguishing concentration of superfine ammonium phosphate extinguishing agent by laser attenuation measurement

Yin Zhiping,Liu Aihua,Pan Renming

Strategic Study of CAE 2008, Volume 10, Issue 7,   Pages 90-95

Abstract:

Extinguishing concentration measurement experiments in the cup burner on two kinds of superfine ammonium phosphate extinguishing agents with different particle sizes had been done by means of filtering and weighing method combined with laser attenuation measurement. The fitting curves of particle concentration with laser attenuation rate and the extinguishing concentration of superfine particle agents had been got. The extinguishing mass concentration of two super fine agents were measured in cup burner. The results showed that when the median particle size of two kinds of extinguishing agents were 6.0 μm and 13.7μm, their laser absorption coefficients were 0.353 m2/g and 0.257 3 m2/g respectively and their mean extinguishing concentration were 32.9 g/m3 and 41.6 g/m3 respectively. Extinguishing ability of the former was 25 % to 30 % higher than that of the latter. The laser attenuation measurement relative errors were less when the former agent was used in experiment.

Keywords： laser attenuation measurement     filtering and weighing method     superfine particle extinguishing agent     extinguishing concentration    

HTML PDF Collect

Title Author Date Type Operation

一种易用的实体识别消歧系统评测框架

辉 陈,宝刚 魏,一鸣 李,Yong-huai LIU,文浩 朱

Journal Article

Disambiguating named entitieswith deep supervised learning via crowd labels

Le-kui ZHOU,Si-liang TANG,Jun XIAO,Fei WU,Yue-ting ZHUANG

Journal Article

A review on cyber security named entity recognition

Chen Gao, Xuan Zhang, Mengting Han, Hui Liu,zhxuan@ynu.edu.cn

Journal Article

A network security entity recognition method based on feature template and CNN-BiLSTM-CRF

Ya QIN, Guo-wei SHEN, Wen-bo ZHAO, Yan-ping CHEN, Miao YU, Xin JIN

Journal Article

A Face Recognition Based on Fusion Features Extraction From Two Kinds of Projection

Zhang Shengliang,Xu Yong,Yang Jian,Yang Jingyu

Journal Article

Learning to select pseudo labels: a semi-supervised method for named entity recognition

Zhen-zhen Li, Da-wei Feng, Dong-sheng Li, Xi-cheng Lu,lizhenzhen14@nudt.edu.cn,davyfeng.c@gmail.com,dsli@nudt.edu.cn,xclu@nudt.edu.cn

Journal Article

A Study on the Essence of Optimal Statistical Uncorrelated Discriminant Vectors

Wu Xiaojun,Yang Jingyu,Wang Shitong,Liu Tongming,Josef Kittler

Journal Article

Both-Branch Fuzzy Decision and Problems on Decision Discernment

Shi Kaiquan,Li Qiqiang

Journal Article

High-Temperature Target Recognition Based on Spectral Radiation Information

Fan Xueliang,Cheng Xiaofang,Xu Jun

Journal Article

DDUC: an erasure-coded system with decoupled data updating and coding

Xiang LI, Yibing LI, Chunrui TANG, Yingsong LI,chunruitang@126.com

Journal Article

A Framework of Knowledge Theory： Toward a Unified Theory of Information, Knowledge and Intelligence

Zhong Yixin

Journal Article

Algorithm Design for Improving Feature Extraction Efficiency Based on KPCA

Xu Yong,Yangjingyu,Lu Jianfeng

Journal Article

Improving entity linking with two adaptive features

Hongbin ZHANG, Quan CHEN, Weiwen ZHANG

Journal Article

Information schema constructs for defining warehouse databases of genotypes and phenotypes of system manifestation features

Shahab POURTALEBI,Imre HORVÁTH

Journal Article

Research on evaluation of extinguishing concentration of superfine ammonium phosphate extinguishing agent by laser attenuation measurement

Yin Zhiping,Liu Aihua,Pan Renming

Journal Article