Wireless body area networks (WBANs) consist of many small low-power sensors, through which users could monitor the real-time parameters of patients' physiology remotely. This capability could improve medical care and the monitoring of patients. WBAN devices typically have limited computing, storage, power, and communication capabilities. These limitations restrict the applications that WBANs can support. To enhance the capabilities of WBANs, the concept of cloud-assisted WBANs has been introduced recently. By using cloud computing technologies, cloud-assisted WBANs can provide more efficient processing of patients' physiology parameters and support richer services. In cloud-assisted WBANs, the data of patients' physiology are stored in the cloud. The integrity of the data is very important because these data will be used to provide a medical diagnosis and other medical treatments. To address the issue of integrity in cloud-assisted WBANs, we propose an efficient certificateless public auditing (CLPA) scheme. A security analysis of our proposed CLPA scheme shows that it is provably secure against two types of adversaries (i.e., a type-I adversary can replace users' public keys, and a type-II adversary can access the master key) in an environment of certificateless cryptography. A detailed performance analysis demonstrates that the proposed CLPA scheme yields better performance over a previously proposed CLPA scheme.

No abstract

Due to the real-time requirement of message in vehicle ad hoc networks, it is a challenge to design an authentication for vehicle ad hoc networks to achieve security, efficiency, and conditional privacy-preserving. To address the challenge, many conditional privacy-preserving authentication schemes using bilinear pairing or ideal tamper-proof device have been proposed for vehicle ad hoc networks in recent years. However, the bilinear pairing operation is one of the most complex cryptographic operations and the assumption of tamper-proof device is very strong. In this article, an efficient location-based conditional privacy-preserving authentication scheme without the bilinear pairing and tamper-proof device is proposed. Compared with the most recently proposed authentication schemes, the proposed scheme markedly decreases the computation costs of message signing and message verification phase, while satisfies all security requirements of vehicle ad hoc networks and provides conditional privacy-preserving.

Both reviews and user-item interactions (i.e., rating scores) have been widely adopted for user rating prediction. However, these existing techniques mainly extract the latent representations for users and items in an independent and static manner. That is, a single static feature vector is derived to encode her preference without considering the particular characteristics of each candidate item. We argue that this static encoding scheme is difficult to fully capture the users' preference. In this paper, we propose a novel context-aware user-item representation learning model for rating prediction, named CARL. Namely, CARL derives a joint representation for a given user-item pair based on their individual latent features and latent feature interactions. Then, CARL adopts Factorization Machines to further model higher-order feature interactions on the basis of the user-item pair for rating prediction. Specifically, two separate learning components are devised in CARL to exploit review data and interaction data respectively: review-based feature learning and interaction-based feature learning. In review-based learning component, with convolution operations and attention mechanism, the relevant features for a user-item pair are extracted by jointly considering their corresponding reviews. However, these features are only reivew-driven and may not be comprehensive. Hence, interaction-based learning component further extracts complementary features from interaction data alone, also on the basis of user-item pairs. The final rating score is then derived with a dynamic linear fusion mechanism. Experiments on five real-world datasets show that CARL achieves significantly better rating predication accuracy than existing state-of-the-art alternatives. Also, with attention mechanism, we show that the relevant information in reviews can be highlighted to interpret the rating prediction.