With the rapid growth of information retrieval technology, Chinese text classification, which is the basis of information content security, has become a widely discussed topic. In view of the huge difference compared with English, Chinese text task is more complex in semantic information representations. However, most existing Chinese text classification approaches typically regard feature representation and feature selection as the key points, but fail to take into account the learning strategy that adapts to the task. Besides, these approaches compress the Chinese word into a representation vector, without considering the distribution of the term among the categories of interest. In order to improve the effect of Chinese text classification, a unified method, called Supervised Contrastive Learning with Term Weighting (SCL-TW), is proposed in this paper. Supervised contrastive learning makes full use of a large amount of unlabeled data to improve model stability. In SCL-TW, we calculate the score of term weighting to optimize the process of data augmentation of Chinese text. Subsequently, the transformed features are fed into a temporal convolution network to conduct feature representation. Experimental verifications are conducted on two Chinese benchmark datasets. The results demonstrate that SCL-TW outperforms other advanced Chinese text classification approaches by an amazing margin.

Remote authentication is a safe and verifiable mechanism. In the Internet of Things (IoT), remote hosts need to verify the legitimacy of identity of terminal devices. However, embedded devices can hardly afford sufficient resources for the necessary trusted hardware components. Software authentication with no hardware guarantee is generally vulnerable to various network attacks. In this paper, we propose a lightweight remote verification protocol. The protocol utilizes the unique response returned by Physical Unclonable Function (PUF) as legitimate identity basis of the terminal devices and uses quadratic residues to encrypt the PUF authentication process to perform a double identity verification scheme. Our scheme is secure against middleman attacks on the attestation response by preventing conspiracy attacks from forgery authentication.

Private data leakage is a threat to current integrity verification schemes of cloud components. To address this issue, this work proposes a privacy-enhancing Structural Integrity Verification (SIV) approach. It is made up of three processes: proof organization, proof transformation, and integrity judgement. By introducing a Merkle tree technique, the integrity of a constituent part of a cloud component on a node is represented by a root value. The value is then masked to cipher texts in proof transformation. With the masked proofs, a structural feature is extracted and validated in an integrity judgement by a third-party verification provider. The integrity of the cloud component is visually displayed in the output result matrix. If there are abnormities, the corrupted constituent parts can be located. Integrity is verified through the encrypted masked proofs. All raw proofs containing sensitive information stay on their original nodes, thus minimizing the attack surface of the proof data, and eliminating the risk of leaking private data at the source. Although some computations are added, the experimental results show that the time overhead is within acceptable bounds.

Infrastructure as a Service (IaaS) has brought advantages to users because virtualization technology hides the details of the physical resources, but this leads to the problem of users being unable to perceive their security. This defect has obstructed cloud computing from wide-spread popularity and development. To solve this problem, a dynamic measurement protocol in IaaS is presented in this paper. The protocol makes it possible for the user to get the real-time security status of the resources, thereby solving the problem of guaranteeing dynamic credibility. This changes the cloud service security provider from the operator to the users themselves. This study has verified the security of the protocol by means of Burrow-Abadi-Needham (BAN) logic, and the result shows that it can satisfy requirements for innovation, privacy, and integrity. Finally, based on different IaaS platforms, this study has conducted a performance analysis to demonstrate that this protocol is reliable, secure, and efficient.