Introduction to the special section on Quality, Reliability and Resilience in Hybrid Information Systems

For decades, the ever-increasing requirement for hybrid information and the continual improvement of hybrid information processing have led to different types of hybrid information systems (HIS). Meanwhile, the quality, reliability and resilience of HIS is the base of its large-scale applications in many research domains. To this end, quality, reliability and resilience of hybrid information system have been recognized as a promising technology to facilitate research in information processing, communication network, internet of things (IoT), etc. Challenges remain unsolved in multiple aspects of quality, reliability and resilience in HIS. First of all, heterogeneous topologies for hybrid information network bring technical challenges in design and management of HIS. Second, in order to manage HIS more effectively, how to measure and analyze the traffic of HIS still needs to be studied. Therein, how to extract various features from hybrid information for content semantic analysis and understanding is an important domain. Third, quality of service necessities for hybrid information causes challenges for big data transmission optimization in HIS. Finally, HIS reliability protection and system resilience enhancement in various environments is one of the bottlenecks which are restricting the large-scale HIS application currently. However, resource constraints of HIS increase reliability and resilience concerns related to their utilization in HIS. All these entire issues need more attention to research today. Therefore, this theme issue was proposed to provide an opportunity for researchers to publish their latest theoretical and technological studies of advanced methods in quality, reliability and resilience of hybrid information processing, and their novel engineering applications within this domain. In this theme issue, 10 out of the 29 submissions were accepted as described below. The first paper titled “Machine Learning-Assisted Signature and Heuristic-based Detection of Malwares in Android Devices”, authored by Irfan Mehmood from Sejong University, Republic of Korea proposed an efficient hybrid framework for detection of malware in Android Apps [1]. In order to avoid inaccurate detection of zero-day attacks and polymorphic viruses provided by currently utilized signature-based methods, the proposed framework considered both signature and heuristic-based analysis. The proposed framework was tested on benchmark datasets by used SVM and KNN classifiers. Experimental results showed improved accuracy in malware detection. The second article titled “A Novel Sentiment Aware Dictionary for Multi-Domain Sentiment Classification”, authored by Vandana Jha from Bangalore University, India, created a sentiment aware dictionary with multiple domain data to solve the domain dependent cost in sentiment analysis which was generated by annotation of corpora in every possible domain of interest before training the classifier [2]. Compared with labeling done by Hindi Sentiwordnet (HSWN), a general lexicon for word polarity, the proposed method was able to label 23-24% more in amount of words in target domain. The comparison between labels assigned by this method and given by HSWN matched with 76% accuracy. The third article titled “Application of rough set theory for NVNA phase reference uncertainty analysis in hybrid information system”, authored by Yang Guohui and Meng Fanyi from Harbin Institute of Technology, China, applied rough set theory to analyze the phase uncertainty of multi-tone stimulus scheme for nonlinear vector network analyzer (NVNA) phase reference frequency response in hybrid information system (HIS) [3]. The proposed scheme provided high resolution and rich intermodulation components around each harmonic for NVNA phase reference standard in HIS. The measured results were in good agreement with the theoretical results and confirmed the validity of traditional work. The fourth article titled “Research on Hybrid Information Recognition Algorithm and Quality of Golf Swing”, authored by Li Jingmei and Zhang Guoyin from Harbin Engineering University, China, proposed a fast golf gesture recognition algorithm of static image and video sequence for the field of sports auxiliary training by used an improved AdaBoost classifier [4]. Experimental results indicated that the golf gesture recognition algorithm of static image had an average recognition time of 2.38ms for each frame in a low failure rate. Moreover, the proposed method was applied in practical with various versions of iPhone and got the recognition speed>30fps and the accuracy 97%.