G2TT
来源类型Project Reports
规范类型报告
Big data analysis : application to environmental research and service
Sung Won Kang
发表日期2017-12-31
出版年2017
语种英语 ; Korean
摘要
The key advantages of Machine Learning analysis using large data are 1) accurate forecast and 2) unknown-pattern finding In this report, we try to make use of these advantages in Environmental Research and Service. This research is composed of three components. First, we apply Machine learning algorithm to environmental research. (2017~19) Second, we accumulate data and algorithms developed in environmental research and combine them with environmental data web crawling algorithm to build environmental machine learning platform(2020~22). Third, we develop public environmental service using these research results and platform(2023~25). In 2017, we developed three machine learning algorithms applied to environment data ? LSTM algorithm estimating hourly find dust pollution, Random Forest/Boosting ensemble algorithm estimating monthly find dust pollution, DNN algorithm estimating intestinal infection case numbers using climate data. Also we applied LDA/Association Rule Learning/word2vec algorithm to online news data and KEI report data, and found that KEI should pay more attention to generic mutation, noise, environmental health, environmental data and specific climate issues like typhoon, severe cold, heavy snow to catch up with public interests represented in online news data.
来源智库Korea Evironment Institute (Republic of Korea)
资源类型智库出版物
条目标识符http://119.78.100.153/handle/2XGU8XDN/198793
推荐引用方式
GB/T 7714
Sung Won Kang. Big data analysis : application to environmental research and service. 2017.
条目包含的文件
文件名称/大小 资源类型 版本类型 开放类型 使用许可
사업2017_07_강성(11961KB)智库出版物 限制开放CC BY-NC-SA浏览
个性服务
推荐该条目
保存到收藏夹
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Sung Won Kang]的文章
百度学术
百度学术中相似的文章
[Sung Won Kang]的文章
必应学术
必应学术中相似的文章
[Sung Won Kang]的文章
相关权益政策
暂无数据
收藏/分享
文件名: 사업2017_07_강성원.pdf
格式: Adobe PDF

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。