Data Science in the Chemical Engineering Curriculum

Data Science in the Chemical Engineering Curriculum
复制标题

化学工程课程中的数据科学

DOI:
--
复制
发表时间:
2019
期刊:
影响因子:
3.5
通讯作者:
T. Duever
T. Duever
中科院分区:
工程技术3区
文献类型:
--
作者:
T. Duever

文献摘要

被引文献

相似文献

随着大量数据的可用性越来越高,属于数据科学术语的方法正在成为化学工程师使用的重要资产。一般来说,需要方法来执行三项任务,即数据管理、统计和机器学习以及数据可视化。虽然有人声称数据科学本质上是统计学,但考虑到前面提到的三项任务,可以清楚地看出,它实际上比单纯的统计学更广泛,而且,数据匮乏时代的统计方法可能还不够。虽然数据科学方法已经有许多成功的应用,但仍然有许多挑战必须解决。例如,仅仅因为数据集很大,并不一定意味着它是有意义的或信息丰富的。从组织的角度来看,缺乏领域知识和缺乏训练有素的劳动力等问题被认为是在组织内成功实施数据科学的障碍。数据科学中采用的许多方法都是化学工程师熟悉的;然而,通常情况下,并非所有进行数据科学项目所需的方法都包含在本科化学工程课程中。解决这一问题的一个办法是调整课程,修改现有课程并开设选修课。其他例子包括引入数据科学未成年人或研究生证书或数据科学硕士课程。
With the increasing availability of large amounts of data, methods that fall under the term data science are becoming important assets for chemical engineers to use. Methods, broadly speaking, are needed to carry out three tasks, namely data management, statistical and machine learning and data visualization. While claims have been made that data science is essentially statistics, consideration of the three tasks previously mentioned make it clear that it is really broader than just statistics alone and furthermore, statistical methods from a data-poor era are likely insufficient. While there have been many successful applications of data science methodologies, there are still many challenges that must be addressed. For example, just because a dataset is large, does not necessarily mean it is meaningful or information rich. From an organizational point of view, a lack of domain knowledge and a lack of a trained workforce among other issues are cited as barriers for the successful implementation of data science within an organization. Many of the methodologies employed in data science are familiar to chemical engineers; however, it is generally the case that not all the methods required to carry out data science projects are covered in an undergraduate chemical engineering program. One option to address this is to adjust the curriculum by modifying existing courses and introducing electives. Other examples include the introduction of a data science minor or a postgraduate certificate or a Master’s program in data science.