Abstract
In complex networks, identifying influential nodes is of great importance for its wide applications. Traditional centrality methods are usually directly based on topological structures of networks, and different centrality methods consider different structural characteristics related to the functional importance. However, in many scenarios, it always exists a complex and nonlinear relationship between the functional importance of a node and its various features including local location, global location, etc., which is hard to be described by one centrality. In order to solve this problem, this paper proposes a framework based on machine learning to measure the importance of nodes in the propagation scenario. This framework first constructs the feature vector of each node based on the existing centrality methods which can reflect nodes’ different topological structures and the infection rate which is an important factor in the propagation scenarios, then labels each node based on the real propagation ability obtained from simulated propagation experiments based on SIR model, last uses seven machine learning algorithms to learn the complex relationship between the real propagation ability of each node and its various structural features. The experimental results in real-world networks show that the classification accuracy of the model based on machine learning is generally higher than that of the traditional centrality methods based on one certain topology.
Highlights
EXPERIMENTAL RESULTS AND ANALYSES we conduct a series of experiments to observe the effectiveness of machine learning model in different conditions compared with traditional centrality methods
Experiments are divided into two parts: (1) the first part is that the training nodes and the testing nodes are from the same network, which can observe the availability of machine learning model; (2) the second part is that the training nodes and testing nodes are from different networks, which can observe the scalability of machine learning model
DICUSSION AND CONCLUSION To identify vital nodes in complex networks, traditional centrality methods are designed directly based on part of network topological structures, which leads to the limitation on performance and flexibility of these methods for identifying influential nodes in propagation scenarios
Summary
The research of identifying influential nodes in complex networks has attracted great interests in recent years because of its great significance on many applications, such as: (1) In economic field, researchers use relevant methods to analyze important countries and regions in the global economic system [1]; (2) In power engineering field, researchers use relevant methods to analyze influential nodes and critical lines in the power transmission network [2]; (3) In advertising field, researchers use relevant methods to find the optimal advertising strategy [3]; (4) In big data analysis field, researchers use relevant methods to find key information in massive data [4]; (5) In life science field, researchers use relevant methods to analyze key neurons in brain neural networks and key proteins in protein interaction networks [5]; (6) In network immunity field, the key target is to find key nodes for. Traditional centrality methods are usually based on topological structures of networks. In many real scenarios such as propagation scenario, the importance of a node is related to its local and global structures and related to other features such as the importance of its neighbors and the infection rates. The common drawback of the traditional centralities is the impossibility of taking all factors into consideration in a single centrality, which results in the limitations on performance and flexibility in real scenario
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
