Introduction to the Special Section on Internet of Behavior for Emerging Technologies

With the maturation of Artificial Intelligence of Things, many countries have promoted the smart city concept to improve citizens’ living quality, encouraging many technology developments on the Internet of Behavior (IoB) that utilizes Internet of Things (IoT) to analyze behavioral patterns. For example, during the epidemic of COVID-19, a face-mask detection system and thermal imaging camera can identify if employees fulfill the standards; the equipment can also check if people keep social distances in public gatherings. Smart Care Systems can utilize IoT to analyze older adults’ behaviors, which understand elders’ living and health conditions or track their diets, heartbeats, and sleep through wearable watches. After collecting and analyzing the data, the system will provide feedback regarding personal health suggestions. IoB is at its initial stage that requires combinations from diverse techniques, such as IoT, big data, and artificial intelligence. These technologies analyze behavioral patterns and benefit enterprises to conduct marketing activities or transfer harmful user behaviors. IoB also requires sensor networks to exchange and share data, which makes it essential to consider the energy consumption issue of the sensors. With the development of large-scale sensors and data collection, it is predictable that there will be more and more IoB applications and framework proposed. IoB needs scholars to involve in-depth researches and present more frameworks that are effective, enabling IoB to achieve real-time behavioral analysis. Given IoB's importance and rich applications, it is a very worthwhile topic of research. For this special issue, the goal is to address more than just IoB algorithms; we hope to explore IoB applications and researches in more areas of study and see how IoB models can take a vast amount of available data and help us uncover undiscovered phenomena, retrieve useful knowledge, and draw conclusions and reasoning.

The first theme of this special issue focuses on “Theories, models, and algorithms for IoB.” The article by Li et al., titled “CoRec: An Efficient Internet Behavior-based Recommendation Framework with Edge-cloud Collaboration on Deep Convolution Neural Networks,” presented an efficient internet behavior-based recommendation framework with edge-cloud integrating with deep CNNs (CoRec). The experimental results illustrated the CoRec significantly outperformed the state-of-art methodologies and cloud-only RNN-based approaches with the real-world datasets. The article by Mohammadi et al., titled “Reinforcing Data Integrity in Renewable Hybrid AC-DC Microgrids from Social-Economic Perspectives,” proposed a new method based on sequential hypothesis testing (SHT) approach to detect the data integrity attack (DIA) on the cyber-physical system environment. The experimental results illustrated the proposed method can detect DIA on renewable units successful and decreased the risk on the operation of electric grids. The article by Fang et al., titled “Travel Time Prediction Method Based on Spatial-feature-based Hierarchical Clustering and Deep Multi-input Gated Recurrent Unit,” presented a travel time prediction (TTP) method based on the spatial-feature-based hierarchical clustering (SFHC) and deep multi-input gated recurrent unit (DMGRU). The experiment results conducted the designed prediction method can achieve the MAPE of 3.3109% and MAE of 2.5658, which outperform various combinations of baseline clustering algorithms. The article by Wu et al., titled “A Privacy Frequent Itemsets Mining Framework for Collaboration in IoT Using Federated Learning,” presented a privacy-preserving data-mining framework based on federated learning and data-mining techniques for joint-venture industrial collaborative activities. The approach has been tested on real industrial datasets and experimental results showed that the approach can achieve high accuracy compared with conventional data-mining techniques while preserving the privacy of datasets. The article by Yuan et al., titled “A MEC Offloading Strategy Based on Improved DQN and Simulated Annealing for Internet of Behavior,” developed an improved offloading algorithm for Mobile Edge Computing (MEC) based on Deep Q Network (DQN) and Simulated Annealing (SA) for Internet of Behavior. The proposed algorithm can maximize the usage of the mobile devices and decrease the energy consumption. The article by Lin et al., titled “A Contactless Authentication System based on WiFi CSI,” integrated the transfer learning technology, Residual Network (ResNet), and the adversarial network to build the contactless authentication system. This system can detect unique human features and remove the environment dynamics under different environments simultaneously.

The second theme of this issue focuses on “New IoB framework for intelligent services and apps.” The article by Mezair et al., titled “Towards an Advanced Deep Learning for the Internet of Behaviors: Application to Connected Vehicle,” developed an advanced deep learning framework for IoB (ADLIoB) and applied it into the connected vehicles. The experimental results illustrated the ADLIoB was superior to the baseline methods both on accuracy and runtime. The article by Mishra et al., titled “Hybrid Mode of Operation Schemes for P2P Communication to Analyze End-point Individual Behaviour in IoT,” designed two hybrid models for Routing Protocol for Low-powered lossy network (RPL) focusing on the aspect of efficient downward communication for Internet of Behaviors. The experimental results inferred that the proposed approach can achieve the lower communication overhead than existing schemes including ARPL, MERPL, and HIMOPD. The article by Li et al., titled “Data Privacy Enhancing in the IoT User/Device Behavior Analytics,” used data anonymization techniques to develop a privacy-preserving solution for both structured data and unstructured data. The article by Zeng et al., titled “Economical Behavior Modeling and Analyses for Data Collection in Edge Internet of Things Networks,” constructed a new framework of leasing edge IoT networks and analyzed the influence of sub-network owners’ dishonest behavior. The experimental results demonstrated the proposed framework can save data collection cost up to 53% compared with existing data collection strategies. The article by Xie et al., titled “Nondeterministic Evaluation Mechanism for User Recruitment in Mobile Crowd-Sensing,” developed a stochastic semi-algebraic hybrid system (SSAHS) model and evaluated the user mobility and behaviors of the mobile crowd-sensing (MCS) systems. Furthermore, a user recruitment scheme based on a nondeterministic evaluation mechanism (NUR) is proposed. Finally, the NUR can collect the historical user data and predict user mobility and behaviors.

All selected articles make remarkable contributions to the evolution of artificial intelligence (AI) and IoT as a foundation for the success of the IoB. We would like to thank all the contributors of this special issue for their remarkable participation and valuable scientific contributions. Most important of all, we deeply appreciate the editor-in-chief, Professor Yunhao Liu, for his kind support for this special issue. We are confident that readers of ACM TOSN and scholars researching in the Internet of Behavior and emerging technologies find this special issue of great interest and benefit.