Open AccessDesign and Implementation of an Early Warning System Based on the Risk Measurement Model
Author(s) -
Shengyu Feng,
Jing Li,
Huan Wang,
Tang Qi,
Liqiong Liu,
Fengnian Yin
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2031/1/012063
Subject(s) - computer science , warning system , random forest , risk management , risk analysis (engineering) , risk assessment , set (abstract data type) , early warning system , data mining , artificial intelligence , computer security , business , telecommunications , finance , programming language
In recent years, there are still some defects in the risk management of China’s inspection and quarantine work, such as the omission of major factors in the conformity assessment of inbound and outbound commodities and the subjective and unscientific assignment of risk indicators, which affect the ability of risk monitoring. Therefore, this paper constructs an inspection and quarantine risk measurement model based on random forest algorithm, and develops a risk early warning system for inspection and quarantine business. Firstly, according to the target requirements, the training data set is extracted from the original rough data set, and the overall data is cleaned and modified. Secondly, the integrated machine learning model is used to select the feature values with better prediction ability. Then, the risk measurement model is constructed based on random forest algorithm and deployed in multi-model parallel mode. Finally, a complete risk early warning system is developed based on the risk prediction model. Through the operation of this system, the abilities of risk analysis and discrimination for inspection and quarantine have been greatly improved, and the comprehensiveness and accuracy of risk prevention and control have been effectively guaranteed.