Abstract Various studies have been conducted to detect network anomalies. However, because anomaly signals are determined by the pattern characteristics using the dataset, the real-time detection problem continues. Even if there is a signal with an attack sign among the constantly transmitted and received signals, the attack cannot be blocked in advance. Moreover, it appears in many places in a distributed denial-of-service (DDoS) attack, so the real-time defense must be the best option. Therefore, it is necessary first to discover the characteristics and elements regarded as abnormal signals to discover anomalies in real time. Finally, by analyzing the correlation between network data and features, extracting the elements of the anomaly, and analyzing the behavior of the extracted elements in detail, we aim to increase the accuracy of the anomaly. In this study, we used Coburg intrusion detection and KDDCup datasets and analyzed the correlation of elements in the dataset using a graph neural network. The calculated accuracy values of the anomaly detection were 94.5% and 98.85%.

Abstract In the past, the emergency responses to disasters such as fire outbreak accidents, accidents that require first aid were slow and not optimal. With human intellect, it was impractical to analyze vast amounts of data regarding the continuity of the numerous environmental changes and the correlation there may be with emergency responses based on past experiences with similar situations. Today, artificial intelligence is presented as a powerful tool to various organizations. Many have already made various attempts to apply this technology as an advisor for emergency response. This research expands on the practicality and effectiveness of utilizing AI as an advisory platform for disaster response based on the big‐data, and also it designs an AI advisor platform for disaster response with big data‐based algorithms. Finally AI advisor function are defined as part of the AI advisor platform, the voice recognition function, natural language processing function, big data coordination function.

Summary The geometric mean is typically used to measure the mean of inflation rate and population fluctuation. It is also used in the description and analysis of singularities and geometric distance spaces. Gradient descent is an integral part of artificial intelligence. In this study, we transform the gradient calculation from conventional quadratic gradient descent algorithms into a root extraction calculation using geometric means. To eliminate the computational complexity of differential operations in gradient calculation and to easily calculate roots using only fundamental arithmetic operations, we introduce the Kai Fang method, the East Asian traditional root extraction method. To do this, we propose a new quadratic gradient descent method based on geometric means and we apply the Kai Fang method with geometric means to create an improved quadratic gradient descent method. The proposed method shows improved computational ease over conventional methods.

Summary Due to the remarkable development of Internet technologies, a great deal of valuable digital data is now transmitted over public networks. To guarantee the security of this data during the transfer process, the authentication of its integrity is extremely important. This paper introduces a robust and secure dual‐watermarking‐based fusion of watermarking, optimization, and a compression method utilizing non‐sub‐sampled contourlet transform (NSCT), redundant discrete wavelet transform (RDWT), and singular value decomposition (SVD). In our method, we first apply the NSCT to a higher entropy sub‐band of the host image. Then, our method uses RDWT‐SVD on higher frequency coefficients of the NSCT image. A similar procedure is followed for both mark images. Finally, an appropriate scaling factor, as obtained by fuzzy inference system, is used to invisibly embed the singular values of both mark data into the host image. Here, any more important mark data are scrambled before the embedding process. The simulation tests reveal that the proposed technique is not only imperceptible and secure but also robust against common attacks. The suggested method has a superior ability to extract hidden information than previous conventional techniques.

Summary Academic discussions on cultural intelligence (CQ) are now paying attention to their potential utilization from various angles. The field of study is expanded not only in business administration but also in psychology, education, tourism, communication, and arts. This is due to the widespread study of global communication competence in multicultural situations because of the deepening of globalization. In this paper, we try to find a way to utilize cultural intelligence model proposed by David Livermore. The aim is to develop education contents for the improvement of cultural intelligence of university students. The target is limited to university students and aims to develop education contents to enhance their cultural intelligence. The main purpose of the study was to measure and analyze the cultural intelligence of university students. For that, the level of cultural intelligence of Korean university freshmen was measured and analyzed. The individual level of the four areas constituting the cultural intelligence was identified, and the difference between the male and female was examined. At the same time, the differences in cultural intelligence were analyzed according to the duration of multicultural contact and experience in the case of foreign language lectures taught by foreigners. Finally, we analyzed how the correlation between the four areas that comprise cultural intelligence is occurring, and as a result, the content and results of this study are expected to be an important foundation for the direction of future development of education contents in universities.

In the fourth industrial revolution, various inter-mediation such as theories, techniques, or implementation would be used.It contains computational intelligence, applied soft computing and fuzzy logic and artificial neural networks, intelligent contents security, model driven architecture and meta-modeling, multimedia contents processing and retrieval, vehicular N/W, big data, intelligence information processing, convergence/complex contents, smart learning, intelligent contents design management, methodology and design theory, intelligent media contents convergence/complex media, social media and collective intelligence, and social media big data analytic.

Summary Evaluation information generated by various users is processed using various requirements and data to make recommendations for solving the problems, and it analyzes satisfaction with the results. Despite people normally utilizes the processed information for decision making, not all information, however, brings positive outcomes to users. There are some users who perceived it negatively. In order to minimize the occurrence of such negative effects, the analysis of various user requirements is essential as well as diversifying user inputs for each requirement. Consequently, the results from individual inputs must be predicted. In the past, since the system relies on a single‐expert system, it is necessary to accept and process various limitations of recommendation and multiple requirements. Therefore, the results of the recommendation also have various problems. In order to solve this problem, this study applied an analytic hierarchy process to multiadvisor configuration. In the proposed system, one or multiple advisors are defined, and after analyzing the predefined requirements, the system accepts only the requirements that can be processed and calculates the individual recommendation results. A recommendation system was going to be studied by learning all situation.

The present study aimed to use the proposed system to measure and analyze brain waves of users to allow intelligent upper limb rehabilitation and to optimize the system using a genetic algorithm. The study used EPOC Neuroheadset for Emotiv with EEG electrodes attached as a non-invasive method for measuring brain waves. The brain waves were measured according to the EEG 10-20 standard electrode layout, which allows measurement of signals from each spot where electrodes are attached based on EEG characteristics. The measured data were added in a database. In the intelligent neuro-fuzzy model, wave transform was used for extracting brain wave characteristics according to user intentions and to eliminate noise from the signals in an effort to increase reliability. Moreover, to construct the option rules of the neuro-fuzzy system, FCM technique and optimal cluster evaluation method were used. Furthermore, the asymmetric Gaussian membership function was used to improve performance, whereas SD and WF divided into left and right sides were used to express the chromosomes. Optimal EEG electrode locations were found, and comparative analysis was performed on the differences based on membership function, number of clusters, and number of learning generations, learning algorithm, and wavelet settings. The performance evaluation results showed that the optimal EEG electrode locations were F7, F8, FC5, and FC6, whereas the accuracy of learning and test data of user-intention recognition was found to be 94.2% and 92.3%, respectively, which suggests that the proposed system can be used to recognize user intention for specific behavior. The system proposed in the present study can allow continued rehabilitation exercise in everyday living according to user intentions, which is expected to help improve the user's willingness to participate in rehabilitation and his or her quality of life.

With the remarkable development of internet technologies, the popularity of smart healthcare has regularly come to the fore. Smart healthcare uses advanced technologies to transform the traditional medical system in an all-round way, making healthcare more efficient, more convenient, and more personalized. Unfortunately, medical data security is a serious issue in the smart healthcare systems. It becomes a fundamental challenge that requires the development of efficient innovative strategies towards fulfilling the healthcare needs and supporting secure healthcare transfer and delivery. This article provides a comprehensive survey on state-of-the-art techniques for health data security and their new trends for solving challenges in real-world applications. We survey the various notable cryptography, biometrics, watermarking, and blockchain-based security techniques for healthcare applications. A comparative analysis is also performed to identify the contribution of reviewed techniques in terms of their objective, methodology, type of medical data, important features, and limitations. At the end, we discuss the open issues and research directions to explore the promising areas for future research.

introduction Share on Introduction to the Special Issue on Recent Trends in Medical Data Security for e-Health Applications Editors: Amit Kumar Singh Search about this author , Zhihan Lv Search about this author , Hoon Ko Search about this author Authors Info & Claims ACM Transactions on Multimedia Computing, Communications, and ApplicationsVolume 17Issue 2sJune 2021 Article No.: 58pp 1–3https://doi.org/10.1145/3459601Published:18 May 2021Publication History 6citation187DownloadsMetricsTotal Citations6Total Downloads187Last 12 Months106Last 6 weeks24 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my Alerts New Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Summary Augmented‐reality (AR) devices allow physicians to incorporate data visualization into diagnostic and treatment procedures to improve work efficiency and safety and reduce cost. They are also used to enhance surgical training. In this study, we implemented an AR application for Botox injections using a face recognition algorithm based on deep learning, and we evaluated the recognition accuracy of this application using 27 participants. The accuracy was around 3 mm for all parts of the facial region. The method of increasing surgical efficiency with AR is accurate enough to be used for surgery and provides great potential for further development.

Summary Each person has unique bio‐information, and this information rarely would be overlapped to other people. Because of this feature, many researchers have been working on a user's identification. However, the problem is that it is not certain if the feature is exactly matched at any time and in any place for the dynamic signal. It is very hard to match bio‐information whenever it is measured; however, all natural signals have an individual pattern. In this paper, it uses an ECG (electrocardiogram) as bio‐information, and it tries to find each pattern, which will be located within the threshold. With the pattern in the threshold, it detects the user's identification. To analyze the patterns, it analyzes them as measured for 120 seconds. Next, it divides them every 1‐2 seconds to 5 seconds. Then, it could recognize the users' identification with this study, and then, finally, the accuracy is 83.3618%.

Summary Context awareness is a necessary technique for providing optimized services to users by recognizing their surrounding environment at a particular time. To provide context‐aware services, a context‐aware middleware is required to detect the changes surrounding the user as well as the processing procedure. However, when the context‐aware middleware is applied to wearable computing, the processing time increases in proportion to the increase in the number of context information due to the lack of processing capacity from the terminal devices. To access the terminal devices, a specific context representation with Resource Description Framework–based triplet is used. Since the triplet consists of three keyword elements that refer to the status of a given situation around the user, retrieving these elements takes O(n3) time complexity with a linear search. To overcome this problem, a hash‐based comparison method is suggested to minimize the response time. The suggested comparison method gives a better performance without searching every single keyword element among the triplet sets. In the experiment, we applied the suggested method to the context‐aware workflow middleware and demonstrated that the proposed method improves the processing time by at least 30% compared with the linear search by enhancing the comparison module in the middleware.

IoT enabled smart car era is expected to begin in the near future as convergence between car and IT accelerates. Current smart cars can provide various information and services needed by the occupants via wearable devices or Vehicle to Everything (V2X) communication environment. In order to provide such services, a system to analyze wearable device information on the smart car platform needs to be designed. In this paper a real time user recognition method using 2D ECG (Electrocardiogram) images, a biometric signal that can be obtained from wearable devices, will be studied. ECG (Electrocardiogram) signal can be classified by fiducial point method using feature points detection or nonfiducial point method due to time change. In the proposed algorithm, a CNN based ensemble network was designed to improve performance by overcoming problems like overfitting which occur in a single network. Test results show that 2D ECG image based user recognition accuracy improved by 1%~1.7% for the fiducial point method and by 0.9%~2% for the nonfiducial point method. By showing 13% higher performance compared to the single network in which recognition rate reduction occurs because similar characteristics are shown between classes, capability for use in a smart vehicle platform based user recognition system that requires reliability was demonstrated by the proposed method.

The modern solutions in Internet technologies like IoT, Cloud, and Pervasive computing, as well as new computation approaches are due to the rapid development of the critical infrastructure and also different methods, providing the secure communication between remoted peers. Recent progress in ubiquitous computing is made possible, thanks to the application of the cutting edge achievements in soft computing, cognitive information systems, and a number of other techniques like ambient intelligence, human-centered computing, bio-inspired computation, etc. Recently, such methods have had a great importance and made it possible to intelligently analyze and perform analytics research in a great amount of complex data (Big Data) as well as manage in secure manner the data transmissions over global communication networks, which also contributing to the development of pervasive and mobile computing (Edge, Fog, and Cloud Computing). The possibility of further development and application of such technology will be dependent on many factors like designing new generation information systems, application of innovative security models for global communication, services management in ubiquitous computing, etc. It may also be dependent on analysis of context around the systems, and the methods of communication, providing a high level of security of the transmitted and aggregated data originating from various places or sensors. These subjects, as well as a number of others, like the innovative and secure applications, as well as the ways of using biologically inspired and cognitive procedures, computational models and secure communication protocols, smart services, or pervasive computing will form the subject of this Special Issue on “Bio-inspired and cognitive approaches in cryptography and security applications” in the Concurrency and Computation: Practice and Experience Journal. To this Special Issue, seven articles of particular interest were selected, as they present the most interesting researches within the subject matter of this special issue. The paper “An Acceleration of Quasigroup Operations by Residue Arithmetic” by P. Kromer, J. Platos, J. Nowakova, and V. Snasel1 presents novel solutions in efficient implementation of quasigroup operations for cryptographic procedures. Residue number systems allow to implement a fast and concurrent realization of addition and multiplication operations, which in this work, were used to accelerate quasigroup operations and create an efficient computational approach to their implementation, designed with respect to the extended instruction sets of modern processors. The paper entitled “Humanics Information Security” by M. Nishigaki2 presents an innovative studies on so-called “humanics information security,” in which the author tries to develop technologies, which ensure both security and usability, by combining the power of humans' cognitive and psychological characteristics. These studies focused mainly on user authentication and captcha enhanced with humanics information security. The image-based authentication, using cognitive aspects such as experiential knowledge and schema, can make the user authentication easy and tolerant. The password authentication using entertainment factors can enhance a user's ability to memorize a stronger password. In the paper entitled “Visual CAPTCHA Application in Linguistic Cryptography” by U. Ogiela, M. Takizawa, and L. Ogiela3 was presented a new idea of linguistic cryptography and linguistic techniques application for data analysis. Linguistic approach may be applied in cryptography, as well as in secret sharing systems implemented in Cloud or distributed computer infrastructure. Described methods allow to distribute data within a different authorized group of secret trustees, based on captcha analysis. The paper “CNN-based Malicious User Detection in Social Networks” by T. Hong, Ch. Choi, and J. Shin4 presents a new techniques proposed for detection of malicious users in social network services. In particular, authors proposed a method of following suggestions from users with less likelihood of committing malicious activities through an information-driven follow suggestion based on a categorical classification of interests using both the images and text of user posts. The images and text are learnt using a convolutional neural network, and the interests are classified into several categories. Users with a large number of posts are defined as certified users, and a database of certified users is established. Users with similar interests were recognized, and the similarity distances between certified users and users are measured, and a follow suggestion is generated to the certified user with the most similar interest. The overall precision during classification of the interest categories was at 79.93%, what indicate a high classification performance. The paper entitled “Design of a Secure and Effective Medical Cyber-Physical System for Ubiquitous Telemonitoring Pregnancy” by G.-Ch. Li, Ch.-L. Chen, H.-Ch. Chen, F. Lin, and Ch. Gu5 presents new solutions, which allow to support selected health care services and personal information management. Proposed solution is based on medical cyber-physical systems (MCPS) and presents a new level of integrated intelligence, characterized by interaction and coordination of computing processes with physical processes. In this paper, authors take advantage of the cloud computing and the high efficiency of MCPS to provide medical monitoring and also to protect pregnancy-related privacy. Identity-based encryption and signature cryptosystem were proposed as the best ways to ensure the security and privacy of health care data in the cloud. In the paper entitled “An MFCC-based Text-Independent Speaker Identification System for Access Control” by J.-Ch. Liu, F.-Y. Leu, G.-L. Lin, and H. Susanto6 was presented a new speaker identification system named mel frequency cepstral coefficients (MFCC), which allow to perform speaker identification for access control. Speaker identification is performed in in frequency domain, and the system uses human auditory filtering model to adjust the energy levels of different frequencies of voice quantified features. Also, a Gaussian mixture model is employed to represent the distribution of the logarithmic features for specific acoustic model, and for accessing a real-world object protected by the speaker identification system, acoustic model is compared with already known acoustic template. Based on the identification result, the identification system will determine whether the access will be accepted or denied. The paper “A Smart Service Model in Greenhouse Environment Using Event-Based Security Based on Wireless Sensor Network” by S. Sivamani, J. Choi, K. Bae, H. Ko, and Y. Cho7 discusses the security solution for the wireless sensor networks in the automated agricultural environment without any human intervention. In proposed solution, the sink node collects the sensor values, organizes the data, and transfers the data to the server, using encryption and private key communication. Secure protocol used in communication and encryption prevents the system from eavesdropping and forging. Some of the malicious events discussed in the papers may be prevented, so finally, data security and confidentiality are increased, what allow the better crop growth in greenhouse environment. We believe that papers included in this Special Issue will have a great impact for future scientific work and also make a contribution to the studies conducted by other researchers and practitioners, who work in the area of intelligent or cognitive systems and advanced computer security. We would like to express our sincere appreciation of the valuable contributions made by all authors. Our special thanks go to Professor Geoffrey C. Fox from Indiana University, Editor in Chief of the Concurrency and Computation: Practice and Experience Journal, for allowing us to publish this Special Issue and for his great support throughout the entire publication process.

Summary Distributed compressive sensing (DCS) usually improves the signal recovery performance of multi‐signal ensembles by exploiting both intra‐ and inter‐signal correlation and sparsity structure. However, the existing DCS had proposed for a very limited ensemble of signals that has only single common information. This paper proposes a generalized DCS (GDCS) framework which can improve sparse signal detection performance given arbitrary types of common information, which are classified into full common information and partial common information after overcoming against existing limitation. Specifically, the theoretical bound on the required number of measurements under the GDCS is obtained. We also develop a practical algorithm to obtain benefits using the GDCS. At the end of this paper, it simply summarizes the potential security issues when it gets all sensing information in a sensor network. Finally, numerical results verify that the proposed algorithm reduces the required number of measurements for correlated sparse signal detection compared to the DCS algorithm. This research lays down the basis for efficient distributed signal detection so that it can improve the detection performance or it can detect the signal reliably when the number of signal observations is limited.

Summary In the smart agricultural environment such as greenhouse or vertical farm, the automation process is performed using the environment sensors to maintain the growth of the crops. Currently, the system rely on the defined rule to perform automation, but the situation can turn catastrophic. With the interruption in the communication, data forging, or eavesdropping, the crops will be rotten and destroyed. Therefore, to maintain the automation without failure, the connection needs to be secure and tampering has to be avoided. In this paper, we discuss the security solution for the wireless sensor networks in the automated agricultural environment without any human intervention. The sink node that collects the sensor values organizes the data and transfer the data to the server, along with the XML encryption and private key mechanism for the communication. The secure protocol is used to make connection using the private key, XML ecryption prevents the system from eavesdropping, and forging. Some of the events discussed in the papers are EVNData, SENSORError, DDoSAttack, and ConnThreat. With the help of the event and the tag information, the data security and confidentiality is increased, for the better automation and crop growth.

ISyRAmI (Intelligent Systems Research for Ambient Intelligence) proposed by IST is an Artificial Intelligence oriented methodology and architecture for the development of Ambient Intelligence (AmI) systems. The ISyRAmI architecture considers the following four modules: Data/Information/Knowledge acquisition; Data/Information/Knowledge storage, conversion, and handling; Intelligent Reasoning; and Decision Support/Intelligent Actuation. Also, Dr. Hoon Ko had presented about ISyRAmI SF, which was involved security model to ISyRAmI in ICWMC2009, ISyRAmI is consists of four modules, that is context allocator, context analyzer, context collector and context detector. Because there are various and many contexts in Ambient Intelligence Environment, contexts are needed to classify according to a characteristic and a purpose. Therefore, we studied contexts classification that will be generated from ISyRAmI.