Authentication email has already been sent, please check your email box: and activate it as soon as possible.
You can login to My Profile and manage your email alerts.
If you haven’t received the email, please:
|
|
There are 22 papers published in subject: > since this site started. |
Select Subject |
Select/Unselect all | For Selected Papers |
Saved Papers
Please enter a name for this paper to be shown in your personalized Saved Papers list
|
1. Classification of security vulnerability exploit codes using large language models | |||
Huang Linhui,He Yongzhong,Yin Min,Li Chao,Hou Lu,Wang Xiaonan,Guo Yaoyao | |||
Computer Science and Technology 26 February 2024 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Software vulnerabilities are the root cause of security risks such as data breaches, system crashes, and network intrusions. Once malicious attackers exploit these vulnerabilities, they can result in significant losses. According to reports from the National Vulnerability Database (NVD), the number of disclosed vulnerabilities is steadily increasing, providing attackers with more opportunities to exploit these vulnerabilities. Consequently, more attack scripts, known as exploit data, are being publicly disclosed. In order to facilitate the use of such data by penetration testers and researchers, relevant individuals have established exploit databases. However, these databases largely rely on manual collection and categorization, making them susceptible to human factors.Therefore, there is a need to employ automated classification methods to effectively manage exploit programs targeting various software and systems. This can enhance management efficiency and reduce associated costs. This article introduces an automated exploit classifier that categorizes exploit information\'s text and code separately. It combines BERT and CodeBERT models along with W2V models to generate corresponding feature vectors. Subsequently, it utilizes models like BiLSTM to construct an automated exploit classifier, achieving effective exploit classification. | |||
TO cite this article:Huang Linhui,He Yongzhong,Yin Min, et al. Classification of security vulnerability exploit codes using large language models[OL].[26 February 2024] http://en.paper.edu.cn/en_releasepaper/content/4762133 |
2. Flexible revocation in ciphertext-policy attribute-based encryption with verifiable ciphertext delegation | |||
Shijie Deng,Gaobo Yang,Wen Dong,Ming Xia | |||
Computer Science and Technology 12 May 2021 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Attribute-based encryption (ABE) is a promising approach to enables fine-grained access control over encrypted data in cloud storage. However, to design a flexible and effective revocation mechanism has always been a tricky problem in ABE schemes, especially for situations where revocation occurs frequently. In this paper, we propose a practical attribute-based access control system by introducing a ciphertext-policy attribute-based encryption (CP-ABE) scheme that allows the trusted authority (TA) to efficiently manage the credentials of data users. The problem of revocation is solved efficiently by exploiting user binary tree. To achieve flexible revocation, our scheme supports both attribute revocation and user revocation to accommodate different revocation needs. Non-revoked users can still decrypt the ciphertext as long as his/her remaining attributes satisfy the access policy associated with the ciphertext. Moreover, verifiable ciphertext delegation is presented to reduce the heavy computation cost brought by frequent revocation. The merits of the proposed scheme are shown by comparing it with the related works. Security analysis and performance discussions further demonstrate the effectiveness of our scheme in cloud systems. | |||
TO cite this article:Shijie Deng,Gaobo Yang,Wen Dong, et al. Flexible revocation in ciphertext-policy attribute-based encryption with verifiable ciphertext delegation[OL].[12 May 2021] http://en.paper.edu.cn/en_releasepaper/content/4754941 |
3. Efficient Decentralized Ciphertext-Policy Attribute Based Encryption With Fine-Grained Revocation | |||
Dong Wen,Yang Gaobo | |||
Computer Science and Technology 12 May 2020 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Ciphertext-policy attribute based encryption (CP-ABE) enables flexible access control for cloud storage. If and only if a user\'s attributes satisfy the pre-defined access policy, which is embedded in the ciphertext, user can decrypt the data. Especially, decentralized CP-ABE is suitable for distributed system due to more flexible attribute management. Nevertheless, the access policy for uploading data to the cloud is explicit, which might lead to privacy leakage. A user\'s privacy can be traced by his/her unique identifier. Existing works either pay little attention to protecting user\'s privacy from attribute-authorities colluding, or have vast computational overheads for enhanced security. In this paper, an efficient decentralized CP-ABE with fine-grained revocation is proposed for cloud storage. To improve privacy protection, user\'s attributes are hide in the access policy, and an anonymous key protocol is constructed to make user\'s be unknown to attribute authorities. Meanwhile, complicated tasks such as pairing and exponential operations in encryption and decryption are outsourced to the cloud. The proposed scheme is friendly to resource-constrained end-users. Even fine-grained attribute revocation used to be complex, our scheme requires only updating part of user\'s secret key and re-encrypting part of ciphertext. Extensive performance analysis shows that the proposed scheme is more efficient and improves privacy, which makes it be more feasible for cloud. | |||
TO cite this article:Dong Wen,Yang Gaobo. Efficient Decentralized Ciphertext-Policy Attribute Based Encryption With Fine-Grained Revocation[OL].[12 May 2020] http://en.paper.edu.cn/en_releasepaper/content/4752051 |
4. A Multiple Images Hiding Scheme Based on Compressive Sensing | |||
WANG Xiqi,PENG Haipeng,LI Lixiang,LI Lixiang,YANG Yixian | |||
Computer Science and Technology 25 December 2019 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:With the rapid development of Internet technology and the rise of the personalized network services, the persons are increasingly inclined to synchronize personal privacy data to network service platforms, such as Baidu Cloud, Tencent Weiyun, Dropbox, etc. Whereas enjoying the convenience of sharing the resources by network services, there are some hidden dangers, for example, the personal privacy data are stolen or maliciously exploited, which brings a great threat to personal information and property security. In order to alleviate the transmission pressure and improve the security, this paper proposed a multiple images hiding encryption scheme based on compressive sensing. In this scheme, the main energy of the multiple plain images is hidden into the carrier image simultaneously, then the compressive sensing is carried out to improve the hiding capacity of the carrier image, ensuring that more encrypted information can be transmitted in a certain storage space. In addition, the proposed scheme will obtain a visually meaningful cipher image if the image is reconstructed directly, which can reduce the risk of further decryption by the eavesdropper and ensure high covertness during the transmission. | |||
TO cite this article:WANG Xiqi,PENG Haipeng,LI Lixiang, et al. A Multiple Images Hiding Scheme Based on Compressive Sensing[OL].[25 December 2019] http://en.paper.edu.cn/en_releasepaper/content/4750318 |
5. An Effective Optimization Method of Measurement Matrix for STP-CS | |||
LI Chong-Xiao,LI Li-Xiang,PENG Hai-Peng,YANG Yi-Xian | |||
Computer Science and Technology 24 December 2019 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Semi-Tensor Product Compressed Sensing (STP-CS) breaks through the limitation of dimension matching between the measurement matrix and the measured data in the traditional compressed sensing (CS), and it greatly reduces the storage space of the measurement matrix. However, the dimension reduction of the measurement matrix leads to the reduction in recovery accuracy. Therefore, in order to improve the recovery effect of the STP-CS, this paper proposes an effective optimization method of the measurement matrix for STP-CS based on the gradient descent, which can reduce the correlation coefficients of the final measurement matrix by optimizing the initial low-dimensional measurement matrix according to the properties of the semi-tensor product and the measurement matrix in STP-CS, thus improving the reconstruction performance of STP-CS. The simulation experiments are conducted to verify the effectiveness of the optimization method proposed in this paper and the experimental results demonstrate that it improves the recovery effect of STP-CS and its recovery effect is very stable. In addition, compared with the optimization method of the measurement matrix in the traditional CS, the proposed optimization method of the measurement matrix in STP-CS can save the storage space, reduce the optimization time and the computational complexity of the measurement matrix optimization, and it has more obvious advantages especially when compressing the high-dimensional data. | |||
TO cite this article:LI Chong-Xiao,LI Li-Xiang,PENG Hai-Peng, et al. An Effective Optimization Method of Measurement Matrix for STP-CS[OL].[24 December 2019] http://en.paper.edu.cn/en_releasepaper/content/4750324 |
6. An Improved Consensus Mechanism Based on Vote Mechanism & Blockchain | |||
Yan Jin,Li Lixiang,Peng Haipeng,Yang Yixian | |||
Computer Science and Technology 23 December 2019 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In the Proof of Stake (PoS) based on the blockchain, the consensus mechanism pseudo-randomly selects a creator within each period of time, and gives the creator the power to produce block. The share of the currency owned by the node, that is, the share of the stake, determines the probability that the node is selected as the representative node. The block created newly must follow the latest blockchain. Therefore, through the displacement of time, most blocks will be added to the same blockchain, making the blockchain continuously grow. However, in the network with poor synchronization, the traditional PoS produces multiple verified representative nodes in each round, which generates multiple blocks. Then the blockchain is easy to produce the bifurcation. Aiming at the bifurcation problem of PoS, we propose an improved voting mechanism based on credit rewards and punishments to vote on multiple blocks. The block is selected by considering the credibility of the creator, the count of the obtained comprehensive votes and the transaction value of the block. Thereby it can ensure the consistency and the fairness of the blockchain network. In the same time, the malicious node is found and voted to be removed in time by counting the number of invalid blocks generated by the nodes. It can ensure the security of the blockchain network. | |||
TO cite this article:Yan Jin,Li Lixiang,Peng Haipeng, et al. An Improved Consensus Mechanism Based on Vote Mechanism & Blockchain[OL].[23 December 2019] http://en.paper.edu.cn/en_releasepaper/content/4750316 |
7. Time-efficient Task Caching Strategy for Multi-server Mobile Edge Cloud Computing | |||
CHEN Weiwei,HAN Limin | |||
Computer Science and Technology 23 April 2019 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Caching tasks to mobile edge cloud (MEC) servers is a proven effective solution for enabling request-duplicate applications on mobile devices (MDs). Moreover, in view of many users in the same proximity are more inclined to require the same computation tasks, reasonable task deployment strategy can reduce task response delay significantly. This promotes us to design an effective task caching policy to save the task computing time for the same computation requests. In this paper, we consider the scenario where multiple mobile devices request computing tasks from a fixed task set to multiple servers, including MEC servers, agent server and remote center cloud server. And the MEC servers can share the cached contents through the proxy server to serve the MDs. Only when there is no cache for a task in MEC servers, MD sends the computation request to remote center server. Our goal is to propose an effective task caching strategy for the MEC servers to minimize the overall response time at the mobile terminal side. To this end, we propose the multi-user multi-server task caching scheme (MUMSC) for the mobile edge computing system. Simulation results show that compared with other caching policies, MUMSC is the the optimal in response time and task hit ratio. | |||
TO cite this article:CHEN Weiwei,HAN Limin. Time-efficient Task Caching Strategy for Multi-server Mobile Edge Cloud Computing[OL].[23 April 2019] http://en.paper.edu.cn/en_releasepaper/content/4748585 |
8. RGB-D Object Tracking and Occlusion Deformation Processing Based on Depth Model | |||
WANG Shuzhen,HE Songhua | |||
Computer Science and Technology 11 April 2019 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:To achieve more accurate RGB-D tracking, robust occlusion and deformation processing, an object tracking method based on depth model is proposed. The algorithm is based on the kernelized correlation filter to satisfy the real-time requirement. The tracking situation is determined by the tracking results of kernelized correlation filter on the color image and depth image. In the unambiguous situation, the linear regression model is adapted to fuse the tracking results. In the ambiguous situation, the relative weight method is used to fuse the tracking results. The Gaussian mixture model is adapted to judge the target state. Partial occlusions and deformations are processed by the region growing method according to the target depth range determined by the model. The algorithm is evaluated using the Princeton data set. The experimental results demonstrate that the tracker achieves more accurate and robust tracking results when the target is partially occluded or deformed. | |||
TO cite this article:WANG Shuzhen,HE Songhua. RGB-D Object Tracking and Occlusion Deformation Processing Based on Depth Model[OL].[11 April 2019] http://en.paper.edu.cn/en_releasepaper/content/4748353 |
9. Research on Recovery Strategy of Protocol Resources in MPLS Multicast Tree | |||
SUN Cong,ZHANG Haiyang | |||
Computer Science and Technology 06 January 2019 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:The MPLS multicast tree does not depend on other protocol clusters, and has the characteristics of simple configuration and self-management. However, in the case of the overload scenario in which the routing device uses the resources of the multicast tree, the link convergence speed is slow and some important services cannot be recovered. The delay strategy can have a good suppression effect on frequent oscillations in the network, and the weight model can give optimal decision for the business recovery process. In this paper, the recovery time and recovery quality in the resource overrun are improved. A general delay-based recovery strategy based on weight calculation is proposed to improve the convergence speed of the multicast tree in the resource overload scenario and improve the service recovery quality. | |||
TO cite this article:SUN Cong,ZHANG Haiyang. Research on Recovery Strategy of Protocol Resources in MPLS Multicast Tree[OL].[ 6 January 2019] http://en.paper.edu.cn/en_releasepaper/content/4746933 |
10. Brand Purchase Prediction based on Time-evolving User Behaviors | |||
Dong Yunqi,Jiang Wenjun | |||
Computer Science and Technology 03 May 2018 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Purchase prediction is a key function in recommendation systems. Existing work usually focus on item-level purchase prediction, which faces two issues of high cost and low accuracy. So in this paper, we study brand purchase prediction, by exploring all possible behaviors which may lead to a purchase on a brand, rather than that on an item.We make three progresses: (1) We intensively analyze various user behaviors and particularly focus on their evolution with time and their diversity on brand. (2) Based on the analysis, we extract several key features that can serve as indicators of users\' purchase intent. Next, we integrate those features into a logic regression based classifier, and construct a brand purchase prediction model. (3) We conduct this model prediction in two scenarios of promotion and non-promotion periods, and we distinguish the different importances of different features. The results show that the proposed model performs well in both scenarios. Moreover, we find that users are more inclined to purchase familiar brand in non-promotion scenarios, while prefer to explore new brands in promotion. | |||
TO cite this article:Dong Yunqi,Jiang Wenjun. Brand Purchase Prediction based on Time-evolving User Behaviors[OL].[ 3 May 2018] http://en.paper.edu.cn/en_releasepaper/content/4744880 |
Select/Unselect all | For Selected Papers |
Saved Papers
Please enter a name for this paper to be shown in your personalized Saved Papers list
|
About Sciencepaper Online | Privacy Policy | Terms & Conditions | Contact Us
© 2003-2012 Sciencepaper Online. unless otherwise stated