New Metrics for Node Importance Evaluation in Occupational Injury Network

• Main Author: Xinbo Ai
• Format: Article
• Language: English
• Published: IEEE 2019-01-01
• Series: IEEE Access
• Subjects: Occupational injury; complex networks; heterogeneous network; node importance
• Online Access: https://ieeexplore.ieee.org/document/8712450/

Complex networks provide a convenient way to model the process of occupational injury occurrence at the system level, and node importance metrics are usually employed to quantify the influence of the factors leading to injuries. However, the traditional metrics such as degree and betweenness are based on the supposition that the network is a homogeneous one and the types of its nodes have to be the same. To describe the injury occurrence, there should be at least two types of nodes, i.e., source nodes and injury nodes. Since this network is heterogeneous in nature, traditional metrics for node importance evaluation are no longer applicable. Hence, we propose two metrics: radial degree and partial betweenness, to quantify the contribution of the source nodes. The former is to describe their induction capabilities, while the latter to depict their control capabilities. The composite score of these two metrics is utilized to evaluate the node importance. The empirical analysis on a total of 438 fatal accident reports in Beijing from 2004 to 2018 showed that our method notably outperformed several state-of-art metrics in evaluating and identifying the crucial nodes.