People's lives have been significantly impacted by social networks, which give a convenient area for them to interact. Every day, a growing number of social networks, such as online social networks, scientific collaboration networks, and airport travel networks, develop and evolve. In social networks, the new connections are created due to the communications between members, or the old connections break due to closed communications, which prompt the network structure to evolve. Furthermore, individual interactions may cause complex collective phenomena in social networks, such as opinion formation, spreading dynamics, and collaborative behaviors, highlighting the essential role of social networks as a complex system. Mining data from social networks and analyzing complicated human behaviors can facilitate understanding the basic mechanism of macroscopic occurrences, discovering potential public interest, and generating timely alerts for collective emergencies. As a result, social network mining has gained a lot of interest as a promising research topic.
Theoretical modeling and data-driven approaches are the two categories of studies on social networks in general. To discover the microscopic dynamical nature of cohesive phenomena, theoretical modeling employs statistical physics, Monte-Carlo simulations, stochastic processes, et al. This kind of research can only interpret phenomena. Data-driven methods utilize data mining, machine learning, and natural language processing to uncover the potential law of online society and forecast future social behaviors. However, the collective phenomena can’t be explained and the uniformly sampling from the entire network may result in bias. Recently, big data created enormous challenges in data processing and human behavior research. Thus, multidisciplinary data analysis methods should be proposed.
The aim of this Special Issue in Frontiers in Physics is to collect original research and review articles related to social networks. We want submissions specifically with multidisciplinary methods for social data mining. The contributions can cover machine learning, information theory, applied mathematics, and computational and statistical physics.
Potential topics include but are not limited to the following:
? Network representation learning;
? Deep learning in social computing;
? Heterogeneous social network mining;
? Behavior analysis on social networks;
? Human sentiment mining and analysis;
? Personalized recommender systems;
? Knowledge graph and its applications;
? Individual interest modeling;
? The essential mechanism of information diffusion and control;
? Modeling the formation and phase transition of collective phenomena;
? Cascade prediction of information propagation.
People's lives have been significantly impacted by social networks, which give a convenient area for them to interact. Every day, a growing number of social networks, such as online social networks, scientific collaboration networks, and airport travel networks, develop and evolve. In social networks, the new connections are created due to the communications between members, or the old connections break due to closed communications, which prompt the network structure to evolve. Furthermore, individual interactions may cause complex collective phenomena in social networks, such as opinion formation, spreading dynamics, and collaborative behaviors, highlighting the essential role of social networks as a complex system. Mining data from social networks and analyzing complicated human behaviors can facilitate understanding the basic mechanism of macroscopic occurrences, discovering potential public interest, and generating timely alerts for collective emergencies. As a result, social network mining has gained a lot of interest as a promising research topic.
Theoretical modeling and data-driven approaches are the two categories of studies on social networks in general. To discover the microscopic dynamical nature of cohesive phenomena, theoretical modeling employs statistical physics, Monte-Carlo simulations, stochastic processes, et al. This kind of research can only interpret phenomena. Data-driven methods utilize data mining, machine learning, and natural language processing to uncover the potential law of online society and forecast future social behaviors. However, the collective phenomena can’t be explained and the uniformly sampling from the entire network may result in bias. Recently, big data created enormous challenges in data processing and human behavior research. Thus, multidisciplinary data analysis methods should be proposed.
The aim of this Special Issue in Frontiers in Physics is to collect original research and review articles related to social networks. We want submissions specifically with multidisciplinary methods for social data mining. The contributions can cover machine learning, information theory, applied mathematics, and computational and statistical physics.
Potential topics include but are not limited to the following:
? Network representation learning;
? Deep learning in social computing;
? Heterogeneous social network mining;
? Behavior analysis on social networks;
? Human sentiment mining and analysis;
? Personalized recommender systems;
? Knowledge graph and its applications;
? Individual interest modeling;
? The essential mechanism of information diffusion and control;
? Modeling the formation and phase transition of collective phenomena;
? Cascade prediction of information propagation.
