ggEDA: Visualisations for exploratory data analysis using tiled one-dimensional graphics and parallel coordinate plots

ggEDA:使用平铺一维图形和平行坐标图进行探索性数据分析的可视化

阅读:1

Abstract

Exploratory data analysis (EDA) involves summarising trends within a dataset to help uncover data quality issues and generate hypotheses. However, identifying relationships between multiple features often requires extensive coding, manual inspection and statistical modelling. Here, we introduce the ggEDA R package, which streamlines multidimensional data exploration by providing two turnkey and complementary visualisation strategies. ggEDA generates interactive parallel coordinate plots (PCPs) well suited for examining large datasets with mostly quantitative features, and introduces tiled one-dimensional plots that more effectively show missingness and reveal categorical relationships in smaller datasets. ggEDA reduces the amount of code and time required to detect multi-feature relationships that may otherwise require statistical modelling or thorough manual review to identify. To make ggEDA visualisations accessible to a wider audience we also developed interactiveEDA, a web app that enables non-programmers to explore and interpret data patterns interactively. ggEDA and interactiveEDA are available at https://github.com/CCICB/ggEDA and https://github.com/CCICB/interactiveEDA respectively.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。