当前位置: X-MOL 学术Cybern. Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Machine Learning to Forecast the Success of Infrastructure Projects Worldwide
Cybernetics and Systems ( IF 1.1 ) Pub Date : 2020-07-28 , DOI: 10.1080/01969722.2020.1798645
Álvaro Herrero 1 , Secil Bayraktar 2 , Alfredo Jiménez 3
Affiliation  

Abstract Governments are increasingly relying on private participation projects and foreign ownership to access technology and capital in infrastructure projects. As a result, the ubiquity of these projects in all regions of the world is a reality that has caught the attention of both managers and scholars. Predicting the final status (success/failure) of these projects in advance is a key element to be taken into account when deciding about participation. To support this kind of decision, the present paper proposes a multidimensional study where a set of heterogeneous classifiers have been applied to forecast the final success of private participation projects. They are applied to a real-life dataset, comprising information from the World Bank about projects all over the world and within four sectors (Energy, Telecommunication, Transport, and Water Sewerage). Classification results are compared under the scope of the sector and host region of the projects. Results show that the predictability of the success of private participation projects depends on the specific industry and region on which the project operates, with projects in the Telecommunication sector and Sub-Saharan Africa exhibiting the highest rates.

中文翻译:

机器学习预测全球基础设施项目的成功

摘要 政府越来越依赖私人参与项目和外资所有权来获取基础设施项目的技术和资本。因此,这些项目在世界各个地区的普遍存在已成为引起管理者和学者关注的现实。提前预测这些项目的最终状态(成功/失败)是决定参与时要考虑的关键因素。为了支持这种决定,本文提出了一项多维研究,其中应用了一组异构分类器来预测私人参与项目的最终成功。它们被应用于现实生活中的数据集,包括来自世界银行的关于世界各地和四个部门(能源、电信、运输、和污水处理)。分类结果在项目所在行业和项目所在地区范围内进行比较。结果表明,私人参与项目成功的可预测性取决于项目运营所在的特定行业和地区,电信部门和撒哈拉以南非洲的项目表现出最高的比率。
更新日期:2020-07-28
down
wechat
bug