Energy Investment Risk Assessment for Nations along China’s Belt & Road Initiative: A Deep Learning Method

In 2013, China proposed the “Belt & Road Initiative” which aims to invest Road” countries so as help them develop their infrastructure and economy. consumes largest part of fossil energy whole world, it is China’s priority consider its supplying security. Therefore, becomes urgent for countries’ facilities. There comes a question: how evaluate overseas investment risk? To answer this question, paper proposes deep learning method assess such risk 50 countries. Specifically, first an indicator system in 6 main factors are separated into 36 sub-factors. This makes use hierarchical convolution neural networks (CNN) encode historical statistics. The structure could CNN handle long statistics more effectively efficiently. Afterward, leverages self-attention layer calculate weights each sub-factor. It be observed that resource potential most important indicator, while “years diplomatic relations” sub-indicator. Finally, we conditional random field (CRF) softmax compute assessment scores country. Based on experimental results, suggests Russia, United Arab Emirates (UAE), Malaysia, Saudi Arabia, Pakistan, Indonesia, Kazakhstan reliable choices investment.