Scale and efficiency of China’s radiation industry based on LightGBM model

Author:

Gong Yu1,Qu Yunhuan1,Xu Ying1

Affiliation:

1. 1 Nuclear and Radiation Safety Center, MEE , Beijing , , China .

Abstract

Abstract In the digitalization of big data information, how to use big data and China’s radiation industry combined have become a key topic of social concern. This paper first constructs LightGBM model by gradient boosting tree and deep neural network. Gradient boosting tree optimizes the convergence and stability of LightGBM model, while the deep neural network improves the accuracy and efficiency of the LightGBM model by updating the weights on each edge. Then the radiation safety management method is proposed according to the danger of the radiation industry, the characteristic data of radiation industry scale and efficiency are obtained through research work, the research indexes are determined, and finally, the data analysis of China’s radiation industry scale and efficiency is conducted by using LightGBM model. Take the medical isotope industry as an example: from 2010 to 2020, the number of nuclear medicine industries increased from 857 to 1148, and the deviation rate of prediction was always maintained at 10%. In terms of radiation industry benefits: the annual growth rate of PET diagnostics, SPECT diagnostics and Nuclide Therapy benefit is 56%, 14% and 5.3%, respectively, the radiation industry has great market demand, and the benefit will continue to grow in the future. This study is a comprehensive and accurate analysis of the scale and benefits of China’s radiation industry and has a guiding reference value for the development and research of China’s radiation industry.

Publisher

Walter de Gruyter GmbH

Subject

Applied Mathematics,Engineering (miscellaneous),Modeling and Simulation,General Computer Science

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3