Table of Content

    15 November 2019, Volume 5 Issue 11
    National New Strategy of Cyberspace Security in the Digital Economy Era
    2019, 5(11):  954-958. 
    Asbtract ( )   PDF (818KB) ( )  
    Related Articles | Metrics
    Content Security Service in the Age of Artificial Intelligence
    2019, 5(11):  959-960. 
    Asbtract ( )   PDF (418KB) ( )  
    Related Articles | Metrics
    AI and Data Privacy Protection: The Way to Federated Learning
    2019, 5(11):  961-965. 
    Asbtract ( )   PDF (1395KB) ( )  
    References | Related Articles | Metrics
    With the tremendous advance in computing, algorithms and data volume, artificial intelligence ushered in the third development climax, and began to gain a foot hold in exploring various industries. However, as the emergence of “big data”, more “small data” or “poorquality data”, and “data silos” exist in industry applications. For example, in the information security realm, it is difficult for enterprises who provide security services such as content security auditing and intrusion detection based on artificial intelligence technology to exchange raw data due to the consideration of user privacy and trade secrets protection. The services between enterprises are independent, and the overall development of cooperation and technology is difficult to make a breakthrough in a short period of time. How to promote greater cooperation on the premise of protecting the privacy of organizations? Will there be any chance for technical means to solve the data privacy protection problems? Federated Learning is an effective way to solve this problem and achieve acrossenterprise collaborative governance.
    Revolution and Challenge: Influence of Intelligent Technology in Media Industry
    2019, 5(11):  966-974. 
    Asbtract ( )   PDF (1333KB) ( )  
    References | Related Articles | Metrics
    In recent years, applications of new technologies such as AI, big data, IoT and VRAR have been resulting in an comprehensive revolution in media industry. These technologies have infiltrated in the whole procedure of content producing including information collecting, checking, processing and users’ feedback. The intelligent distribution technology has brought the new model and platforms. In the future, customized information distribution service may be offered by digital “personal steward”. New technologies are also bringing many new risks. Adhering to ethics norms is vital for the prevention of risks in the time of AI.
    Knowledge Graph and Deep Learning Empowering Content Security
    2019, 5(11):  975-980. 
    Asbtract ( )   PDF (785KB) ( )  
    References | Related Articles | Metrics
    The early idea of knowledge graph originated from the Semantic Web proposed by Tim Berners Lee, the Father of World Wide Web. It aims to use the graph structure to model the relationship and knowledge between the world. Deep learning is derived from the study of artificial neural networks. It uses deep neural networks to learn knowledge from massive data. Its advantage is that it can automatically learn feature from massive data instead of manual feature engineering. Integrating the method of deep learning into the application of knowledge map is one of the current research hot spots. In the fields of automated knowledge acquisition, knowledge representation learning and reasoning, and largescale graph mining and analysis, deep learning and knowledge graph have made a lot of progress. For the content security field, the knowledge graph can effectively improve the retrieval efficiency of content security, enhance the understanding and interpretability of text content, and help content security to move toward the era of knowledge intelligence.
    Negative Sentiment Detection in Intelligent HumanMachine Dialogue
    2019, 5(11):  981-987. 
    Asbtract ( )   PDF (2081KB) ( )  
    References | Related Articles | Metrics
    Artificial intelligence makes the application of intelligent humanmachine dialogue system more and more widely, but the dialogue system often contains negative sentiment information such as vulgarity, violence and rumors, which has a bad influence on the dialogue system. Therefore, it is very important to detect negative sentiment in the dialogue system. Aiming at the negative emotion content in the intelligent humanmachine dialogue system, a negative sentiment detection model based on deep learning is proposed. This model can effectively capture text semantics and context information by using pretraining word vector and BiLSTM. Compared with the traditional dictionary matching algorithm, it greatly reduces the dependence on the dictionary, and also can intelligently identify similar negative sentiment expression forms, which makes detecting negative sentiment in the dialogue system more effectively and plays an important role in purifying the dialogue system.
    A Survey of Image Captioning Technology
    2019, 5(11):  988-992. 
    Asbtract ( )   PDF (1054KB) ( )  
    References | Related Articles | Metrics
    Image captioning is an important task in the field of computer vision and natural language processing. It has a wide and important application value in our life and entertainment, intelligent transportation and helping people with visual impairment. Compared with other perception tasks such as image classification and object detection, image captioning is a higher level and more complex cognitive task, which has a great significance to help analyze and understand images. In this paper, we aim to give a comprehensive overview of the existing image captioning techniques. Here we discuss the data sets and evaluation metrics commonly used in image captioning, as well as the performances, advantages and limitations of existing image captioning techniques.
    A Spammer Detection Method Based on Social Network Structure
    2019, 5(11):  993-999. 
    Asbtract ( )   PDF (1428KB) ( )  
    References | Related Articles | Metrics
    In the social network, after some harmful accounts are detected and intercepted, new spammers will be derived to continue to spread negative comments and rumors, which seriously damages the interests of the public. In the past, spammer detection methods were based on linguistic features and nonverbal behavioral features (such as writing habits). Although some success has been achieved, some clever puppet masters can easily disguise their language and behavioral features to evade detection. It is difficult to guarantee the performance of these detection methods. However, the social structure between users is not fully exploited and utilized in social networks. In this paper, a spammer detection method based on social network structure is proposed, to transform the identification into a similar subgraph matching problem. The method proposed in this paper is carried out on the Sina Weibo dataset, and the experimental results prove the effectiveness of the proposed spammer detection method.
    AI Security—Research and Application on Adversarial Example
    2019, 5(11):  1000-1007. 
    Asbtract ( )   PDF (3155KB) ( )  
    References | Related Articles | Metrics
    With the rapid development of AI (artificial intelligence), the number of AI systems and applications grows explosively. AI has been closely linked to numerous people and brings great convenience to their life. Meanwhile, AI also leads to big challenges in the cyber security area. Some malicious fraudsters take advantage of AI to attack internet systems especially in the field of captcha generation. The antiknowledge map captcha based on the adversarial example technology is proposed, which fused the natural language processing technology and adversarial example generation technology, and thus increase the robustness to attacks and safeguard the security environment of internet.
    Secure Face Verification with Deep Learning: Status and Challenges
    2019, 5(11):  1008-1012. 
    Asbtract ( )   PDF (1330KB) ( )  
    References | Related Articles | Metrics
    As an uncontrolled, userfriendly biometric verification technology, face recognition is widely used in industries such as public security, finance and personal mobile devices. Face recognition is a typical representative of success computer vision applications for real world problem. Thanks to massive datasets, increasing hardware computing power, and rapid development of deep neural networks technology, deep learning based face verification has made great progress in recognition performance, and has achieved much better performance than the traditional methods. However, the face recognition technology with high recognition accuracy has obvious security holes under malicious forged identity attacks. Techniques such as face security and face antispoofing are attracting more and more attention from both academic and industry. We will first introduce some of the most recent mainstream recognition and antispoofing techniques, and then introduce several challenges in deep learning based face antispoofing, such as insufficient supervision, training data bias, lack of generalization across cameras and lack of representative datasets.
    Public Opinion Situational Awareness in Cyberspace
    2019, 5(11):  1013-1019. 
    Asbtract ( )   PDF (6996KB) ( )  
    References | Related Articles | Metrics
    With the development of the Internet, the influence of online public opinion on the society has become more and more prominent, and it has been widely concerned by party and government organs and enterprises. In order to effectively enhance the ability of discovery, research and evaluation of network public opinion, this paper studies the monitoring and analysis technology of network public opinion based on situational awareness. Firstly, this paper puts forward the analysis model of network public opinion situational awareness and discusses the technical method of its realization. Then, through the practical application case, the construction method of the indicator system and the realization scheme of the technical system in the network public opinion situational awareness system are elaborated in detail. Based on the integration of multiple public opinion sensation elements, the case system constructs an effective quantitative index system, realizes the state evaluation of network public opinion from a macroscopic point of view, and predicts the development trend of network public opinion under certain conditions. Comprehensive awareness and realtime warning of online public opinion information and events were implemented. Based on this, we can further study the mechanism of linkage with the business and realize the business closed loop of network public opinion monitoring.v
    Study on Identifying and Governing Internet Hate Speech
    2019, 5(11):  1021-1026. 
    Asbtract ( )   PDF (905KB) ( )  
    References | Related Articles | Metrics
    he popularization of Internet has created convenience for the dissemination of hate speech and brought about certain negative social effects particularly within the EU. Due to the differences in history and culture of different countries, there are also great differences in the understanding and governance of hate speech in the international community. EU countries advocate combating the spread of hate speech in legal form, but the United States does not accept the concept. Meanwhile, the impact of regulatory activities on freedom of expression is also difficult to assess. Thats the reason why its difficult to control hate speech in the international area. Based on the international communitys understanding and governance practice of hate speech, this paper puts forward the constitutive factors of hate speech as a criterion. Then combining the practice of governing hate speech on overseas Internet platforms and in the EU region, this paper analyzes the specific situation of governing hate speech in China.
    The Technical Principles Behind the Toutiao Recommendation System
    2019, 5(11):  1027-1032. 
    Asbtract ( )   PDF (1294KB) ( )  
    References | Related Articles | Metrics
    Since launched in 2012, Toutiao has been committed to connect people and information by artificial intelligence, make the distribution of high quality, abundant information more efficient and accurate, create values for users. Toutiao has now become one of the largest information platforms in China, which own more than 100 vertical channels, such as science and technology, sports, health, food, education, agriculture, rural areas and farmers, Chinese style, the NBA, etc, supporting multiple sort of genres of online information, such as graphic and text, atlas, small video, short video, short articles, live video streaming, mini program, etc. To meet the users demands for obtaining valuable information in such an exploding information era, Toutiao has taken the lead in the field of recommendation systems and made significant achievements through a lot of explorations and practice, such as applying the largescale cluster streaming computing, natural language processing, deep learning to content recommendation and quality evaluation processing, text and image recognition, video understanding and recommendation, multimedia creation and visual platform, etc.
    Research on a Security Intelligent Bearing Network Architecture for Smart City
    2019, 5(11):  1033-1039. 
    Asbtract ( )   PDF (3314KB) ( )  
    References | Related Articles | Metrics
    tThe new generation of information technology is driving vigorous development of smart city construction. Under the new situation of the rapid development of smart city construction, the operation and management of the city put forward new demands on the diversity, continuity and elasticity of business support networks. By analyzing the current research status of smart city bearing network architecture, a new smart city safe intelligent bearing network—Single Network Multiplain Network Architecture (hereinafter referred to as smart network) is proposed. The network architecture has the characteristics of security, reliability, high scalability and key business classification. It provides an ideal choice for solving the network architecture of smart city basic bearing network by integrating MPLSVPN, SDN and other network technologies.
    Solution of InterDomain Identity Authentication Based on Bridge CA
    2019, 5(11):  1040-1043. 
    Asbtract ( )   PDF (1366KB) ( )  
    References | Related Articles | Metrics
    In the E-government extranet, the basic network environment is complex, and there are many independent CA operating organizations which form their own certificate application domains. How to realize the unified identity authentication across many certificate domains involves two problems, one is how to label the identity uniquely, the other is how to identify the identity. In a certificate domain, identity is identified by a certificate. Across many certificate domains, there are multiple trust anchors (the starting point of trust). The problem of trust between trust anchors should be solved first, and then the uniqueness of identity information represented by the certificate can be explained. The second is to solve the problem of certificate validation, that is, the problem of identity authentication. This paper proposes a solution of inter-domain identity authentication based on bridge CA to solve the problem of multi-CA trust through bridge CA and realize crossdomain identity authentication through interdomain gateway, and presents the implementation process of crossdomain identity. Under the premise of not changing the existing application scenarios, this scheme can well solve the identity authentication problem of crossdomain identity by building a new trust chain and using a new method.K
    Cybersecurity Talent Cultivation in the View of Cybersecurity Law
    2019, 5(11):  1044-1048. 
    Asbtract ( )   PDF (7071KB) ( )  
    References | Related Articles | Metrics
    With the generalization of emerging cybersecurity risks and the emergence of new technologies and applications, the fundamental value of cybersecurity talent is gradually highlighted, which makes countries continuously deepen the understanding of “human factors”. As an essential law on cyberspace in our country, Cybersecurity Law and its supporting systems reflect the consideration of “human factors” at the social and national levels from the aspects of publicity and education, support and promotion, personnel empowerment and reverse restriction. In the everlasting progress of the toplevel design into the implementation of the system, China must maintain the stability of legislation and policies about “human”, mobilize the enthusiasm of all participants in the society, take multiple means to cultivate cybersecurity talents, and cultivate cyber security culture that can effectively support the construction of a strong cyber country.