Due to the multi-modality and diversity of the data collected, Sensor Data Analysis approaches can differ significantly in terms of the chosen analysis methods. Commonly found solutions include: 1) methods that look to detect the presence of specific engagement cues/events such as directed gaze, back-channels, valence, smile [34, 60], 2) supervised classifiers where the labels come from human annotators [15, 41, 64], and 3) deep-learning  and deep reinforcement learning [53, 61] approaches for engagement estimation. The deep-learning methods are relatively newer methods in HRI, motivated by the idea that the traditional machine learning methods are not equipped to deal with high-dimensional feature space, require expert engineering, and always rely on data annotation. While the first kind of methods are relatively straight-forward to implement, they are limited to the detectable cues, which are few and possibly affected by confounding factors. Even though supervised classifiers are one of the widely used methods, since engagement is a highly subjective construct, there is the problem of generalization and accuracy of such models since they are modeled in a specific context and the labels are provided by multiple annotators. We must also note that not many studies actually report the annotation protocol. Lastly, the latest deep learning approaches suffer from the lack of interpretability/explainability of results and require an abundance of data.
The learning task lies at the heart of the activity and requires the children to interact with maps such as those shown in Fig. 3 via touch-screens, as shown in Fig. 1. A small humanoid robot, acting as the CEO of a gold-mining company reiterates the problem by asking the participants to help it collect the gold by connecting the gold mines with railway tracks, while spending as little money as possible. The participants collaboratively construct a solution by drawing and erasing tracks that connect pairs of goldmines, and submit it to the robot for evaluation (one of the two optimal solutions is shown in Fig. 3).
The two views provide complimentary functionality and, therefore, in order to make informed decisions, the team members need to communicate. While in the figurative view, one can build and erase tracks, in the abstract view, one can view the cost of every track ever added, access previous solutions and their costs, and bring back a previous solution.
Maria is a QA Lead with 14 years of experience working with various multinational companies based in the US and Europe. While in the US, she also completed a master's degree in business. Maria started her career as a Manual QA intern and then worked her way up to senior and lead roles. She is passionate about providing quality software to clients as well as sharing her QA knowledge and experiences with aspiring QA professionals.
Michael is a Software QA Engineer who has experience at Indeed.com and Sure App. He worked in sales prior to getting into QA, and is now passionate about improving the quality of software and sharing his knowledge with others. He likes to help people who want to start their career in the technology industry.
Provincial Health Services Authority (PHSA) improves the health of British Columbians by seeking province-wide solutions to specialized health care needs in collaboration with BC health authorities and other partners.
Internet of Things (IoT) is the utmost assuring framework to facilitate human life with quality and comfort. IoT has contributed significantly to numerous application areas. The stormy expansion of smart devices and their credence for data transfer using wireless mechanics boost their susceptibility to cyberattacks. Consequently, the cybercrime rate is increasing day by day. Hence, the study of IoT security threats and possible corrective measures can benefit researchers in identifying appropriate solutions to deal with various challenges in cybercrime investigation. IoT forensics plays a vital role in cybercrime investigations. This review paper presents an overview of the IoT framework consisting of IoT architecture, protocols, and technologies. Various security issues at each layer and corrective measures are also discussed in detail. This paper also presents the role of IoT forensics in cybercrime investigation in various domains like smart homes, smart cities, automated vehicles, and healthcare. The role of advanced technologies like artificial intelligence, machine learning, cloud computing, edge computing, fog computing, and blockchain technology in cybercrime investigation is also discussed. Lastly, various open research challenges in IoT to assist cybercrime investigation are explained to provide a new direction for further research.
The security of a computer system encompasses various methods and techniques that safeguard all kinds of resources from illegitimate access. Resources may include hardware, software, and data, whereas illegitimate access may include unauthorized usage or damage to resources. In IoT systems, security aspects focus on architecture, the security model of every device, bootstrapping, network security, and application security . Security architecture demonstrates the various system components involved in ensuring the security of an IoT device. The security model of each device focuses on the implementation of security methods and criteria along with the management of various applications. Network security deals with the reliable functioning of IoT. Online application security is all about the authentication of various things on the network for communication and exchange of data. Network security is highly dependent on the internet, which is an anxious media of data exchange and leads to a large possibility of data stealing. The deployment of IoT is dependent on the internet and computer networks. Consequently, it is affected by all security issues related to computer networks as well as the internet. Before using IoT devices, all stakeholders should analyze the associated risks related to the security and privacy of the user information. Accordingly, more sophisticated security policies must be designed by governing organizations.
In , Aggarwal et al. discussed a security prospectus exclusively from a privacy perspective, whereas other security challenges in IoT platforms are not discussed. Said  discussed various IoT architectures along with research issues. In this survey, only challenges faced in physical security and privacy are explored. Moreover, security issues are discussed without giving any viable solutions. Perera et al.  elaborated that security and privacy challenges are handled at the middleware level in the IoT framework and at different layers. In this survey, security is expressed as a normal issue and the authors did not pay any special attention to the research in the field. Granjal et al.  presented an in-depth review of the different security mechanisms and protocols of the time for communication among smart devices. The authors also highlighted the available scope of research. However, on the negative side, the authors did not consider all security standards in their survey but focused on only a few. Sicari et al.  reviewed security from three different angles: security requirements, privacy, and trust. Under security requirements, the authors explored the issues related to access control, confidentiality, and authentication. The biggest drawback of this work is the inadequacy of the categorization of research activities in the IoT security paradigm. Abomhara and Køein  reviewed the security threats along with the security and privacy research challenges in their paper. They stressed research issues like interoperability of diverse IoT devices and authorization.
Mahmoud et al.  surveyed IoT security principles. The authors also presented various security issues along with corrective measures. The need for advanced technologies to tackle hardware, software, user identification, and wireless communication issues is also discussed. Pescatore and Shpantzer  presented the viewpoint of people actively involved in the research of IoT security issues along with the future prospects in the field. They also highlighted that IoT developers should focus more on security issues instead of other ICT systems. Gil et al.  reviewed various technologies and security models in the context of data-related challenges. The authors impressed upon the collaboration of social networks and IoT and introduced a new concept of the Social Internet of Things (SIoT). IoT security is discussed but the concept of cybersecurity in IoT is not touched. Muhammad et al.  discussed the various possible attacks in IoT systems. The authors also highlighted the security and privacy challenges faced in the IoT environment by the various sensor nodes. In this survey, the requirements of secure end-to-end communication among smart devices using efficient encryption and authentication methods are suggested. Vignesh and Samydurai  reviewed the three-layered architecture of IoT comprised of the application, network, and perception layers, along with the different types of security threats at these layers. They explained the effect of wireless signals, movement of IoT in the external environment, and the dynamism of the network model as the major challenges at the perception layer. At the network layer, the major highlighted challenges are DoS and Man-in-the-Middle attacks. The major issue that persists at the application layer is the variety of application policies.
Smart devices and applications in various areas of IoT make human life more comfortable, but they also make IoT systems more vulnerable to cyberattacks. These devices and applications are connected to the internet, which creates new opportunities for cybercriminals to enter the IoT environment. Cybercriminals can enter an IoT system through routers and can damage it in many ways. Although several security mechanisms are available in IoT, advanced technologies like artificial intelligence (AI), machine learning (ML), neural networks (NN), blockchain technology, fog computing, and edge computing are playing a major role to handle cyberattacks and helping to control cybercrime [167, 168]. Authors in  discussed in brief the various kinds of security threats in an IoT environment. The need for a dynamic and quick system to safeguard the IoT systems against cybercrime is impressed upon. The authors proposed a hybrid system to detect cyberattacks using AI and ML in a cloud computing environment. Both types of attacks, i.e., at the device level and the network level, can be detected with this model. According to the authors, it is considered by the security experts that AI and ML provide very powerful security mechanisms as even future attacks may be predicted based on past IoT attack data. Consequently, this system does not wait for the occurrence of attacks but it can predict them in advance. The main limitation of the system is that it can work only with standard data formats for prediction. ML provides solutions to DoS attacks, eavesdropping, spoofing, and privacy leakage in an IoT environment . The authors in  presented a multilayer architecture to associate the various devices within IoT to make them accessible throughout the network at all times. To deal with the security issues of end nodes and to provide more credible services, a novel framework using NN was proposed. According to this framework, security issues need to be tackled in each layer of the IoT architecture. Each end node configured using this framework will have the potential to self-monitor and recover after any unwanted event/attack. In the proposed framework, a NN-based adaptive model was used for the automatic recovery of the nodes. In , the authors presented an artificial neural network (ANN) approach to control distributed denial of service (DDoS) attacks. The ANN was tested in a simulated IoT environment. The results obtained with the proposed technique were found to be 99.4% accurate, and this technique is capable of identifying numerous DDoS/DoS attacks. The authors in  highlighted that the incorporation of blockchain in IoT systems has numerous benefits. The distributed architecture of blockchain reduces the risk of failure of data storage nodes. Thus, it leads to more secure data storage in the IoT environment [173, 174]. The concept of data encryption is used by blockchain for data storage in the IoT environment; so, there are less chances of storing damaged data in things . The augmentation of blockchain with IoT also helps to prevent unauthorized access, data loss, and spoofing attacks . Various challenges in IoT along with the workable solutions administered by the blockchain technology are discussed below in Table 8. 2b1af7f3a8