Special issue on Big Data Security and Intelligent Data in Clouds (BDS‐IDC)

Abstract

Recently, the demands of big data combining with cloud computing have enabled dramatically growing of intelligent data in various fields. Data security issues have been considered one of the most critical aspects for the implementations of big data and cloud computing. Big Data Security and Intelligent Data in Clouds (BDS-IDC) have been immerse in people's daily life and are promised to have a great growth in the future, such as smart data analysis, smart big data, sensitive-oriented information protection, and cloud-computing security. Multiple protection approaches and combined security schemes as well as intercrossed cloud platforms are becoming common requirements for many big data and cloud users. Considering the next generation of big data executions, it is expected to integrate big data with approaches of intelligent data in cloud systems. The main challenges of achieving this aim will go across most emerging technologies from cyber security and high performances including effective security mechanism against old and new threats, complex system architectures, computing resources allocation optimizations, distributed heterogeneous computing, energy-saving issues, and using smart data in cyber physical systems. This special issue presents a number of outstanding papers researching in the latest explorations and studies in Big Data Security and Intelligent Data in Clouds. We encourage audiences who are interested in high-performance pattern recognition to read following work. Li et al. 1 developed algorithms and investigated various classifiers to determine the authenticity of short social network postings, an average of 20.6 words, from Facebook. This work presented and discussed several experiments using a variety of classifiers. The research was one of the first works that focused on authorship authentication in short messages, such as posting on social network sites. Moreover, Yibin Li et al. 2 designed an approach to minimize the necessary calculations and memory space. The authors developed Field-Programmable Gate Array (FPGA)–based parallel architecture to apply Partial Dynamic Reconfiguration (PDR), which enables the remote configurations for the number of processing units. According to the experimental results, the detection precision could reach a miss rate less than 1.97% and a false positive rate of 1%. A reliable broadcast scheme is also a crucial issue in the target field. Xia et al. 3 investigate a novel reliable broadcast scheme that explored the advantages of lost data piggybacking. The scheme allowed all the vehicles to piggyback some received packets cooperatively to help other vehicles to recover the lost packets. The simulation results showed that the proposed cooperative offloading scheme could achieve much higher broadcast reliability and lower propagation delay, in comparison with existing solutions. Furthermore, the implementation of cloud computing is broad in multiple domains. Dou et al. 4 introduce an energy-aware dynamic VM scheduling method for QoS enhancement in clouds over big data to address the above challenge. This method consisted of 2 main VM migration phases to reduce service prices and execution time, which had been evaluated by experiments. In addition, Leng et al. 5 presented a balanced RDF graph partitioning algorithm for storing massive RDF data on cloud. The authors first devise a modularity-based multilevel label propagation algorithm to partition RDF graph roughly and then use a balanced K-mediods clustering algorithm for final k-way partitioning. This special issue also addresses the aspect of combining video with social networks. Cui et al. 6 proposed novel video recommendation algorithm based on the combination of video content and social network. The proposed algorithm consisted of the trust friends computing model and video's quality evaluation model. The trust friends computing method took into account similarity between users, interaction between users, and the active degree of a user. Son et al. 7 introduced a new conditional proxy re-encryption (CPRE) scheme, namely, the CPRE for Mobile Cloud (CPRE-M), which utilizes the back-end cloud to the extreme extent so that the overhead of terminals is drastically reduced. The scheme outsources a significant amount of computation overhead caused by re-encryption key generation, condition value change, and decryption, to the cloud. The proposed scheme allows users to verify the correctness of outsourced computation under refereed delegation of computation (RDoC) model. The simulation results show CPRE-M outperforms its existing alternatives. Cahyani et al. 8 examine the extent to which data acquisition from Windows phone is supported by 3 popular mobile forensics tools. The effect of device settings modification and alternative acquisition processes on the extraction results is also examined. The results show that current mobile forensic tool support for Windows Phone 8 remains limited. In this sense, the logical acquisition support was more complete in comparison to the physical one. Furthermore, enabling flight mode and disabling location services are recommended to limit data alteration during the acquisition process. Existing information security incident handling strategies may not be adequate as cloud data are generally virtualized, geographically distributed, and ephemeral, presenting both technical and jurisdictional challenges. Ab Rahman et al. 9 present an integrated cloud incident handling and forensic-by-design model. Afterwards, the authors validate the model using a set of controlled experiments on a cloud-related incident. Three popular cloud storage applications were deployed, namely, Dropbox, Google Drive, and OneDrive. This study demonstrates the utility of the model for organizational cloud users to undertake incident investigations. To improve the security of the microblog network, Liu et al. 10 proposed a defending scheme against malicious URL spreading in microblog networks with Hub nodes. After a node found a new malicious URL, it edits a warning message about the malicious URL. If the normal node gets malicious message, it sends private message to the Hub node and update its blog article. Malicious URL messages spread rapidly in the networks due to the influence of Hub nodes. So the security of entire microblog networks can be improved against malicious URL without increasing the network load. Experiments show that the scheme can effectively defend against malicious URL in the any scale of microblog networks. Man et al. 11 introduced a new social network access control model using logical authorization language, named as RuleSN, which can be efficiently used in cloud systems. The model provides high performance of authorization expressiveness and flexibility that can effectively describe relations of User to User, User to Resource, Resource to Resource, and attributes of users and resources. The paper elaborates the formal definitions of the RuleSN model, then the model's authorization specification and verification policies are described, and finally, the implementation, application, and expressiveness of the model are discussed. Recommender systems have shown great potential to address information overload problems, namely, to help users find interesting and relevant objects within a huge information space. To achieve more accurate recommendation, Jianxun et al. 12 investigate a recommendation algorithm INBIw, which improves on the original weighted network-based inference (NBIw) by introducing a tunable parameter β to depress the influence of high-degree nodes. To evaluate the recommendation performance of INBIw, ranking position rate and hitting rate are calculated. The results of experiments based on MovieLens data set show that the INBIw outperforms previous methods, including the global ranking method, collaborative filtering, network-based inference, and NBIw. The high-energy–consuming petroleum equipment in oilfield represents a high proportion in the total energy consumption of petroleum industry. Zhao et al. 13 analyzes the energy consumption constitutions in different production processes, focusing on the running state and energy-saving countermeasure of the high-energy–consuming mechanical equipment. The proposed model provides concurrent data processing for shortening the production cycle and the Fuzzy Weighting Subspace Clustering algorithm to implement the equipment comparability, applying the big data on the running status statistics and applying cloud computation analysis on the corresponding energy-saving countermeasure. Wang et al. 14 present a novel virtual network embedding (VNE) algorithm to increase revenue and utilization of substrate network as well as improve acceptance fairness of virtual networks. To this end, they present a virtual topology pretransformation mechanism leveraging reusable technology. Then they model the problem as an integer linear programming problem and solve the VNE problem with a discrete particle swarm optimization–based algorithm. An incentive convergence mechanism is proposed to reduce mapping complexity, which is used to accelerate convergence and to save more bandwidth. The results prove that the proposed method improves existing algorithms in terms of physical resource utilization, acceptance fairness, revenue/cost ratio, and searching efficiency. Li et al. 15 model the collaboration relations in the regional enterprise cluster as a generalized social collaboration network and explore the dynamic growth process of the Facilities Collaboration Network (FCN) for different strategies of facilities selection, including the random selection and the balanced selection. Using performance indexes such as network size, the distribution of node degree and act degree, clustering coefficient, the average shortest distance, and the number of n-cliques, they analyze the characteristics of these strategies for cloud manufacturing. Based on these characteristics, they propose 2 mechanisms for self-optimization, including the dynamic weighing of facilities and the concentrated processing of successive subtasks in the process. Finally, they analyze the mechanisms' effects on the characteristics of FCN and the performance in manufacturing. Anomaly detection plays a crucial part in identifying unforeseen attacks for network and information security. To address this issue, Li et al. 16 propose a Multilayer Anomaly Detection approach, which extracts and combines features from different network layers. To reduce redundancy and noise, an algorithm called RanPF is designed by applying principal components analysis (PCA) into random forest (RF) algorithm. RanPF uses features selected by PCA to decide the height of every tree in RF and provides a method to select which features for tree nodes to use. To obtain high-quality features, they adopt an attribute learning mechanism. Naive Bayes is used to characterize the attribute information. A series of experiments conducted on 2 real-life datasets demonstrate that the approach outperforms the state-of-the-art methods in terms of detection rate and false alarm rate. Tao et al. 17 investigated the cloud resource allocation for data access with noncooperative game based on a Gauss–Seidel iteration method. They proposed a repeated game scheme, which can approximately achieve the near Pareto-optimal flow allocation among the vehicles. Considering the vehicles' irrational behavior, a punishment strategy was designed to prevent the vehicles from behavior deviation. The analysis based on these models lays a theoretical method foundation on cloud resource allocation process. The validity of the modeling and the accuracy of the analysis were verified through extensive simulations. In the past decade, there has been a surge of interest in exploring the properties of scale-free networks. Two interesting properties have attracted much attention: assortativity and community structure. However, these properties have been studied separately in either theoretical models or real-world networks. Jiang et al. 18 show that both properties are highly related in scale-free networks. First, due to the positive degree of assortativity, assortative networks present high embeddedness. This leads to the high density of communities. Second, because of the negative degree of assortativity, disassortative networks exhibit great hubs in communities, which results in the high compactness of communities that nodes can reach each other via a few hops. Finally, as a result of the equal attractiveness of nodes in neutral networks, a big portion of links act as community bridges. Thus, neutral networks display sparse and less compact communities. Li et al. 19 investigate the selfish load balancing problem in mobile distributed crowdsourcing networks. This work leverages the d-choice method based on Ball and Bin theory for effective balancing under limited information and the proportional allocation scheme for selfish load balancing, maintaining good load balancing property among selfish users. The results show that, chance-choice outperforms several existing algorithms, comparing with proportional allocation scheme, it could decrease the load gap between the maximum and the minimum in system by 50% to 80% and reduce the overhead complexity from O(n) to O(1) comparing with the Max-weight Best Response algorithm. This special issue is composed of a collection of articles, which provide recent advances in some fields related to Big Data Security and Intelligent Data in Clouds. More precisely, the manuscripts investigate different topics including forensic in cloud-of-things devices, encryption in mobile cloud environments, defense against malicious URL spreading in microblog networks, social network access control model in cloud computing, dynamic resource sharing algorithms, load balancing for cloud computing, cloud-based audio/video streaming techniques, energy-aware scheduling in clouds, sensor network security in cloud computing, as well as real world applications related to industry, manufacturing, and vanets. We hope that the readers could benefit from the articles presented in this special issue and would contribute to these exciting research areas. The guest editors of this special issue wish to express their sincere gratitude to all of the authors who contributed to this issue. We extend our thanks to the reviewers for their hard work and their feedbacks. Likewise, we wish to express our gratitude to the Editor-in-Chief Geoffrey C. Fox for his valuable help in arranging this special issue and for giving the authors the opportunity to publish their work in Concurrency and Computation: Practice and Experience journal. Lastly, we wish to thank the journal's staff for their assistance.