Discussion of 'Event history and topological data analysis'.

Discussion of 'Event history and topological data analysis'.
复制标题

DOI:
10.1093/biomet/asab023
复制
发表时间:
2021-12
期刊:
影响因子:
2.7
通讯作者:
Ombao H
Ombao H
中科院分区:
数学2区
文献类型:
--
作者:
Chung MK;Ombao H

文献摘要

参考文献

相似文献

虽然拓扑数据分析已经存在了几十年,理论发展很好,但它仍然存在许多统计和计算问题。由于这些原因,它尚未成为数据科学家的标准工具。作者指出,由于拓扑特征的异质性,直接将现有的统计模型应用于持久同源性是困难的。在过去的十年中,拓扑数据分析的统计发展一直集中在通过变换或平滑将异构特征转化为同构结构化数据。因此,将生存分析技术应用于拓扑特征的生灭过程的想法是非常有趣的。作者成功地阐明了事件历史方法与拓扑特征寿命之间的联系,并引发了许多新的有趣的问题。
Although topological data analysis has been around for many decades with well-grounded theoretical development, it still suffers from numerous statistical and computational issues. For these reasons, it has not yet become a standard tool for data scientists. The authors point out the difficulty of directly applying existing statistical models to persistent homology due to the heterogeneous nature of topological features. The statistical development in topological data analysis in the last decade has been focused on making heterogeneous features into homogenous structured data by transformations or smoothing. Thus, the idea of applying survival analysis techniques to the birth and death process of topological features is very intriguing. The authors succeeded in elucidating the connection between event history methods and the lifetime of topological features, and the paper has stimulated many new interesting questions.
DOI: 10.1214/17-aoas1119
发表时间: 2018-09
期刊: The annals of applied statistics
影响因子: --
作者:
Wang Y;Ombao H;Chung MK
通讯作者: Chung MK
DOI: 10.1371/journal.pone.0182184
发表时间: 2017-08-15
期刊: PLOS ONE
影响因子: 3.7
作者:
Li, Yanjie;Wang, Dingkang;Wang, Yusu
通讯作者: Wang, Yusu
DOI: 10.1214/15-aoas886
发表时间: 2016
期刊: The annals of applied statistics
影响因子: --
作者:
Bendich P;Marron JS;Miller E;Pieloch A;Skwerer S
通讯作者: Skwerer S
DOI: 10.1109/tmi.2020.2967451
发表时间: 2020-06
影响因子: 10.6
作者:
Huang SG;Lyu I;Qiu A;Chung MK
通讯作者: Chung MK
DOI: 10.1162/netn_a_00091
发表时间: 2019-01-01
影响因子: 4.7
作者:
Chung, Moo K.;Lee, Hyekyoung;Solos, Victor
通讯作者: Solos, Victor