Is a Dataframe Just a Table?

Is a Dataframe Just a Table?
复制标题

数据框只是一张表吗?

DOI:
--
复制
发表时间:
2019
期刊:
PLATEAU@UIST
影响因子:
--
通讯作者:
Yifan Wu
Yifan Wu
中科院分区:
--
文献类型:
--
作者:
Yifan Wu

文献摘要

被引文献

相似文献

数据查询是数据库和数据科学的核心。然而,这两个社区似乎有不同的概念和用例。因此,查询语言的设计者和用户都不同意核心抽象--结构(数据科学)和表(数据库)--以及操作是否相同。为了从PL-HCI的角度研究这种差异,我们确定了表和框架提供的基本规则,以及表和框架的编程经验如何改变。我们表明,数据结构推动程序员查询和存储他们的数据在不同的方式。我们希望案例研究能够澄清困惑,消除错误信息,增加两个社区之间的交叉授粉,并确定开放的PL-HCI问题。
Querying data is core to databases and data science. However, the two communities have seemingly different concepts and use cases. As a result, both designers and users of the query languages disagree on whether the core abstractions – dataframes (data science) and tables (databases) – and the operations are the same. To investigate the difference from a PL-HCI perspective, we identify the basic affordances provided by tables and dataframes and how programming experiences over tables and dataframes differ. We show that the data structures nudge programmers to query and store their data in different ways. We hope the case study could clarify confusions, dispel misinformation, increase cross-pollination between the two communities, and identify open PL-HCI questions.