摘要
缅甸政治与军事冲突所带来的潜在风险,极有可能对我国多方面造成深远而巨大的影响。缅甸危机蕴藏着政治、经济、文化、军事以及人口等各类不确定因素,同时又受到难以确定的各类内外环境影响,对其危机的风险预测及研判充满困难。因此,要在不确定性的信息条件下去识别、预测和评价缅甸危机对我国带来的风险,须经过一系列审慎的科学评估。本文基于人工智能思想,充分运用数据挖掘技术,采用XGBoost与马尔科夫链等算法,力求超越经典范式的因果推断、相关性分析、机制解释以及依赖回归模型的传统量化预测手段,为缅甸危机可能带给我国的影响构建了一种可供科学预测和评估的工具,也做出了平衡经验研究与人工智能的学术尝试。
Most probably,the Burmese political and military conflicts will bring potential risk with multiple tremendous and far-reaching impacts on all walks of life in our country.It is of extreme difficulty to predict,analyze or judge the risk of the Burmese crisis with regard to uncertain factors in politics,economy,culture,military and population under the influence from both home and abroad.Therefore,a series of prudent scientific evaluations must be carried out in order to recognize,predict and evaluate its risk to our country in the condition full of uncertain information.In this paper,a scientific predictive and evaluative model is constructed to analyze the impacts of the Burmese political and military conflicts to our country.This construction is based on AI theories,with full application of data excavation technology and the method of XGBoost algorithm and Markov chain.It goes beyond the classical causal inference,relevance analysis,mechanism explanation,and the traditional measure of quality prediction relying on regression model,and is an academic attempt in balancing experience research and AI.
作者
秦博
徐浩铭
Qin Bo;Xu Haoming(School of Foreign Languages,UESTC,Chengdu,Sichuan,610000;Department of Teaching and Research of Modern Science and Technology,Party School of Chengdu Municipal Committee CPC,Chengdu,Sichuan,610000)
出处
《成都大学学报（社会科学版）》
2022年第1期104-115,共12页
Journal of Chengdu University (Social Sciences)
