基于数据驱动的野生生物保护的应用研究
Applied Research Based on Data-Driven Wildlife Conservation

野生生物是地球环境的重要组成部分，然而受到非法野生生物贸易的影响，野生生物的种类和数量正面临着严重的威胁和衰退。为了显著减少非法野生动物贸易，本研究对美国野生生物保护法规体系进行了分析，通过多元岭回归模型，对当地野生生物保护种类数、公众对非法野生动物贸易的认知度、公众对加强执法的支持率、公众每年对野生动物保护组织的捐款数、野生生物保护机构年投资额、野生生物保护政策数量、野生生物保护地数量及面积、当地猎人证数量这9个重要指标进行分析，评估他们对非法野生动物年贸易量、年度GDP、当地犯罪统计数、当地人口数等4个政府重点关注因素的影响。并采用灰色预测构建了数据驱动下，野生生物保护的预测模型。最终得到针对性的政策和投资可以在促进社会经济发展的同时，大大降低非法野生生物贸易规模及当地犯罪率。
Wildlife is an important part of the Earth’s environment, yet wildlife species and populations are facing serious threats and declines due to the impact of illegal wildlife trade. In order to significantly reduce illegal wildlife trade, this study analyzed the US wildlife conservation regulatory system by using multivariate ridge regression modeling for the number of local wildlife conservation species, public awareness of illegal wildlife trade, public support for increased law enforcement, number of public donations to wildlife conservation organizations per year, annual investment in wildlife conservation organizations, number of wildlife conservation policies, nine important indicators, namely the number and area of wildlife reserves and the number of local hunter permits, were analyzed to assess their impacts on four government-focused factors, including the annual trade in illegal wildlife, the annual GDP, the number of local crime statistics, and the number of local population. And a data-driven, predictive model for wildlife conservation was constructed using gray prediction. The final result is that targeted policies and investments can make significant progress in reducing illegal wildlife trade and crime rates while promoting socioeconomic development.