Social engineering attacks are becoming serious threats to cloud service. attackers could get Cloud service custom privacy information or attack virtual machine images directly. Existing security analysis instruments difficult quantify the social risk, resulting in invalid defense guidance for attacks. In this article, a risk framework is proposed based on user profiling. The provides pathway quantitatively calculate possibility of being compromised by and potential loss, so as effectively...

In this paper, an interactive multitask learning method for Chinese text sentiment classification is proposed.Here, the classic BiLSTM + attention CRF model used to obtain full use of interaction relationship between tasks, and it simultaneously solves two tasks emotional dictionary expansion classification.The proposed divides into primary task subtask, adopts Enhanced Language Representation with Informative Entities (ERNIE) as representation task.Then, through maximum pooling layer fully...

With cybersecurity situation more and complex, data-driven security has become indispensable. Numerous data exists in textual sources analysis is difficult for both analyst the machine. To convert information into structured further automatic analysis, we extract cybersecurity-related entities propose a self-attention-based neural network model named entity recognition cybersecurity. Considering single word feature not enough identifying entity, introduce CNN to character which then...

A key challenge for the semantic Web is to acquire capability effectively query large knowledge bases. As there will be several competing systems, we need benchmarks that objectively evaluate these systems. Development of effective in an emerging domain a challenging endeavor. In this paper, propose requirements driven framework developing base systems (SW KBSs). make two major contributions. First, provide list SW KBS benchmarks. This can serve as unbiased guide both benchmark developers...

We describe our method for benchmarking Semantic Web knowledge base systems with respect to use in large OWL applications. present the Lehigh University Benchmark (LUBM) as an example of how design such benchmarks. The LUBM features ontology university domain, synthetic data scalable arbitrary size, fourteen extensional queries representing a variety properties, and several performance metrics. can be used evaluate different reasoning capabilities storage mechanisms. demonstrate this...

In the current enterprise network environment, multistep targeted cyber-attacks with concealment and advanced characteristics have become main threat. Multisource security data are prerequisite of detection. However, these characters heterogeneity semantic diversity, existing attack detection methods do not take comprehensive sources into account. Identifying predicting intention from heterogeneous noisy can be meaningful work. this paper, we first review different fusion mechanisms...

The EPC Class-1 Generation-2 specification(Gen-2 in brief) is widely accepted as the standard for RFID tags under grant number ISO18000-6c. However, there are two problems view of its security authentication protocols. First all, no unified requirements, such which threats should be protected against. Secondly, acceptable means to guarantee security, provable methods not applicable without using encryption function or hash function. In this paper, we study requirements presented current...

Most of the previous work on threshold-cryptography-based distributed CA concentrates initial systems configurations and concrete protocols design, ignoring efficiency effectiveness key management service during its operation, always assuming that there are honest nodes to carry out faithfully. This paper focuses developing a selection mechanism in MANETs with selfish nodes, dynamically select coalition carrying threshold optimally system operation. First, we formulate dynamic problem as...

In order to obtain high quality and large-scale labelled data for information security research, we propose a new approach that combines generative adversarial network with the BiLSTM-Attention-CRF model from crowd annotations. We use find common features in annotations then consider them conjunction domain dictionary feature sentence dependency as additional be introduced into model, which is used carry out named entity recognition crowdsourcing. Finally, create dataset evaluate our models...

The ability to reason over large scale data and return responsive query results is widely seen as a critical step achieving the Semantic Web vision. We describe an approach for partitioning OWL Lite datasets then propose strategy parallel reasoning about concept instances role on each partition. partitions are designed such that can be reasoned independently find answers sub goal, when unioned together, complete set of found goal. Our has polynomial worst case time complexity in size...

Wireless networking technologies based on the IEEE802.11 series of standards fail to authenticate management frames and network card addresses suffer from serious vulnerabilities that may lead denial service attacks, this paper proposes a client puzzle defense mechanism resist two resource depletion type DoS attacks: authentication association in proposed mechanism, we ask solve send by access point, could only be solved brute force. The difficulty degree easily adjusted point. In our...

As an emerging network, Delay Tolerant Mobile Sensor Network (DTMSN) is susceptible to be attacked because of its limited resources and intermittent connectivity. Though authentication technology usually used prevent network from attacks, it a challenging work in delay tolerant circumstance. In this paper, node mutual scheme based on Physical Unclonable Function (PUF) proposed for DTMSN. We take advantage the physical characteristics chips with PUFs sensor nodes cloning. Analysis indicates...

Current intrusion detection systems are mostly for detecting external attacks, but the “Prism Door” and other similar events indicate that internal staff may bring greater harm to organizations in information security. Traditional insider threat methods only consider audit records of personal behavior failed combine it with business activities, which miss happened during a process. The authors operators' correctness performance propose process mining based system. system firstly establishes...

In order to find suitableWeb services from a large collection of Web services, automatic support is needed filter out relevant according some criteria specified by the user. real business scenarios constraints on types input and output parameters are often not sufficient. Rather one wishes specify relationships parameters, interaction pattern non-functional properties ofWeb services. Therefore, there need for more expressive goal specification language. Current techniques matchmaking...