Search

Abstract: In this paper, we define a new homomorphic signature for identity management in mobile cloud computing. A mobile user firstly computes a full signature on all his sensitive personal information (SPI), and stores it in a trusted third party (TTP). During the valid period of his full signature, if the user wants to call a cloud service, he should authenticate him to the cloud service provider (CSP) through TTP. In our scheme, the mobile user only needs to send a vector to the access controlling server (TTP). The access controlling server who doesnʼt know the secret key can compute a partial signature on a small part of userʼs SPI, and then sends it to the CSP. We give a formal secure definition of this homomorphic signature, and construct a scheme from GHR signature. We prove that our scheme is secure under GHR signature. [Copyright &y& Elsevier]

At present, various types of code vulnerabilities lead to a huge workload of detection. Although multi-party cooperation detection methods can be applied to improve efficiency, common crowdsourcing models are difficult to guarantee the privacy of code and identity. We design a code and identity privacy vulnerability crowdsourcing detection system ConGradetect. The system is built on a blockchain and solves code privacy issues, identity privacy issues, and reward preemption issues during the crowdsourcing detection process of vulnerabilities. In ConGradetect, users have a dynamic fake identity, which ensures that the identity of the user is not exposed, thereby avoiding the problem of information exposure caused by data accumulation. In the meantime, ConGradetect uses a local code granulation tool to perform off-line code cutting for task publishers, and uses proxy re-encryption to allocate task code blocks to ensure that the complete original code is only known to the code owner. When the task is received, a trusted key conversion is performed using smart contract. Further, we propose a method for anonymous verification of the binding relationship between task and user to prevent reward preemption and ensure that task performers can obtain corresponding legal benefits. Finally, we run the system on the Ethereum blockchain and test its performance. After testing, the feasibility of ConGradetect is proved. [ABSTRACT FROM AUTHOR]

Neural Topic Modeling has attracted significant attention from the Natural Language Processing community due to its black-box inference property and has made some progress. However, existing approaches have three main drawbacks: (1). hard to integrate external semantic knowledge to improve topic quality; (2). the posterior collapse issue of the Variational Auto-Encoder (VAE) harms the topic diversity; (3). cannot capture the semantic scopes of extracted topics. To address these limitations jointly, we model topics with a mixture of von Mises-Fisher distributions to incorporate word semantic knowledge and propose the Spherical Embedded Topic (SET) model based on the Wasserstein Auto-Encoder. Experiments have been conducted on three widely used text corpora, and the results show that SET obtains more coherent and diverse topics, outperforming the state-of-the-art approaches. Moreover, the ability of SET to capture semantic scopes of topics is also validated by our experiments. [ABSTRACT FROM AUTHOR]