Kumar Ankur, Misra Subhas Chandra, Chan Felix T S
Industrial Management Engineering Indian Institute of Technology, Kanpur, Kanpur, Uttar Pradesh, India.
Department of Decision Sciences, Macau University of Science and Technology, Taipa, Macao.
Expert Syst Appl. 2022 Dec 30;210:118628. doi: 10.1016/j.eswa.2022.118628. Epub 2022 Aug 22.
COVID-19 pandemic has given a sudden shock to economy indices worldwide and especially to the tourism sector, which is already very sensitive to such crises as natural calamities, terrorist activities, virus outbreaks and unwanted conditions. The economic implications for a reduction in tourism demand, and the need to analyse post-COVID-19 tourism motivates our research. This study aims to forecast the future trends for foreign tourist arrivals and foreign exchange earnings for India and to formulate a model to predict the future trends based on the COVID-19 parameters, vaccinations and stringency index (Government travelling guidelines). In the study, we have developed artificial intelligence models (random forest, linear regression) using the stacked based ensemble learning method for the development of base models and meta models for the study of COVID-19 and its effect on the tourism industry. The architecture of a stacking model consists of two or more base models, often referred to as level-0 models, and a -model that combines the predictions of the base models, and is referred to as a level-1 model (Smyth & Wolpert, 1999). The results show that the projected losses require quick action on developing new practices to sustain and complement the resilience of tourism per se.
新冠疫情给全球经济指标带来了突如其来的冲击，尤其是对旅游业而言，该行业对自然灾害、恐怖活动、病毒爆发及不利状况等危机本就极为敏感。旅游业需求下降所带来的经济影响，以及分析新冠疫情后旅游业情况的必要性，促使我们开展此项研究。本研究旨在预测印度外国游客入境人数和外汇收入的未来趋势，并构建一个基于新冠疫情参数、疫苗接种情况和严格指数（政府出行指南）来预测未来趋势的模型。在该研究中，我们运用基于堆叠的集成学习方法开发了人工智能模型（随机森林、线性回归），用于构建基础模型和元模型，以研究新冠疫情及其对旅游业的影响。堆叠模型的架构由两个或更多基础模型（通常称为0级模型）以及一个结合基础模型预测结果的模型（称为1级模型）组成（史密斯和沃尔珀特，1999年）。结果表明，预计的损失需要迅速采取行动，制定新的措施来维持和增强旅游业自身的复原力。
