NEWS.md

In linkspotter: Bivariate Correlations Calculation and Visualization

linkspotter

linkspotter 1.3.0

Computation

• add targetVar parameter for multiBivariateCorrelation() and linkspotterComplete() to consider only couples including targetVar
• add maxNbBins parameter for MaxNMI in multiBivariateCorrelation() and linkspotterComplete()
• print info from multiBivariateCorrelation()
• better deal with the parameter includeNA
• bug fix on EFdiscretization() for weird cases
• better handle non informative variables and couples of variables
• remove tidyr and dplyr deprecated functions
• give up Hmisc dependency

Visualization

• add a new feature: hide isolated nodes
• add a new feature: flip the bivariate plot
• add busy spinner from shinybusy package
• use ggplot2 for all bivariate plots

linkspotter 1.2.0 (initial CRAN release)

Computation

• Calculation of several correlation matrices corresponding to different link coefficients
• Clustering of variables using an unsupervised learning
• Supervised discretization of one or a couple of variables.

Visualization

• Visualize the links using a graph (the variables correspond to the nodes and the links correspond to the edges)
• Show the distribution of each variable using its histogram or barplot
• Visualize a link between a couple of variables using scatter plots, box plots, etc.

linkspotter documentation built on July 23, 2020, 5:08 p.m.