[JCR] Predicting the distribution of COVID-19 through CGAN

Predicting the distribution of COVID-19 through CGAN—Taking Macau as an example

Abstract: Machine learning (ML) is used to predict distribution of COVID-19 distribution in urban areas. This study utilizes a conditional generative adversarial network (CGAN) to predict COVID-19 hotspots based on urban texture and business formats. By establishing a relationship between urban elements and COVID-19, machines can automatically predict epidemic hotspots. Using Macau as an example, the method correlates urban texture and business hotspots with new COVID-19 clusters. Different samples yield varying prediction accuracies. Results show: (1) CGAN accurately predicts COVID-19 distribution with over 70% accuracy. (2) Predicting using urban texture and hospital/station POI data achieves over 60% accuracy in Macau. (3) The method also predicts other at-risk areas, aiding urban epidemic control. [SCITIP]

DOI:  https://doi.org/10.3389/fdata.2023.1008292

CITE: Zheng L, Chen Y, Jiang S, Song J and Zheng J (2023) Predicting the distribution of COVID19 through CGAN—Taking Macau as an example. Front. Big Data 6:1008292. doi: 10.3389/fdata.2023.1008292


作者的话

1. 心得分享

本期推文旨在分享 2023 年第一篇发表在 ESCI 期刊《Frontiers in Big Data》的论文《Predicting the distribution of COVID-19 through CGAN—Taking Macau as an example》(通过 CGAN 预测 COVID-19 的分布——以澳门为例)。阅读原文链接可以转跳至出版社网页。浏览器按一下一键翻译中文可查看大概意思。可能有的人会讲这个论文过时了吧,去年 8 月投稿的,今年 1 月录用。我有什么办法,杂志社又不是我开的,政策也不是我能预测的。这个题材我估计以后也很少会再写了。至于普适性问题,传播路径差不多的传染病跟空间分布的关系还是有一定的意义的。

[JCR] Predicting the distribution of COVID-19 through CGAN

2. 简单讲讲

机器学习 (ML) 是一种广泛应用于数据预测的创新方法。使用 ML 预测 COVID-19 分布对于城市安全风险评估和治理至关重要。本研究使用条件生成对抗网络(CGAN)构建一种通过城市肌理和商业业态预测 COVID-19 热点分布的方法,并建立城市元素与 COVID-19 之间的关系,以便机器可以自动预测城市中的流行热点。以澳门为例,利用该方法确定澳门的城市肌理和商业热点与新的疫情热点群之间的相关性。不同类型的样本提供了不同的流行病预测准确度。结果显示如下:(1) CGAN 可以准确预测 COVID-19 的分布区域,准确率可以超过 70%。(2) 通过城市肌理和医院、车站 POI 数据预测 COVID-19 分布的结果最好,在澳门不同区域的实验中准确率超过 60%。(3)所提出的方法还可以预测城市中可能存在 COVID-19 风险的其他区域,有助于城市疫情防控。

TABLE OF CONTENTS

Abstract

1. Introduction

2. Research methodology and data sources

3. Results analysis and discussion

4. Conclusion and outlook

Data availability statement

Author contributions

Funding

Acknowledgments

Conflict of interest

Publisher’s note

Supplementary material

Footnotes

References

3. 文章插图

[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN

4. 后记

《Frontiers in Big Data》似乎是一本新刊,我个人理解,他是走在风口上的热点,大数据、机器学习的专题非常多。短短几年时间能进入 ESCI,未来可期。投稿时间线:

[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN

目前在出版社官网、researchgate 都是全文可以获取的,不需要机构登录,因为是 OA 开源。也欢迎与我们的 RG 帐号互粉。

[JCR] Predicting the distribution of COVID-19 through CGAN

同时也十分感谢学术媒体在推特上面对本文的推介:

[JCR] Predicting the distribution of COVID-19 through CGAN
[JCR] Predicting the distribution of COVID-19 through CGAN

有人会讲,Frontiers、MDPI、Hindawi 这种垃圾的 OA 出版社就不要发了,玷污学术人生。第一,未来是否选择从事学术科研,这也并非是本人唯一选择。第二,MUST 我所在的学院 FA 认 SCI、SCIE、SSCI、ESCI、EI、AHCI 为博士生毕业的一级论著,我读的是 MUST 又不是那些张口闭口 OA 辣鸡的高档单位。任何一个出版社期刊都有好的文章差的文章,各花入各眼,自己做论文问心无愧就得啦,有空可以看看小木虫、发表记还有万维书刊网,OA 不是有钱就能发,不少期刊近年来换了主编门槛、拒稿率都在飙升。以前我也很不屑,觉得发本地的不检索的期刊没卵用哦,但别人的学院看啊,认啊,我也投稿过,的确录稿有难度,那又能咋的,每个人都有自己的选择,只不过在合适的时间做出合适的选择罢了🤔

正文完
更多资讯,敬请关注「走访建筑」微信公众号😘
post-qrcode