summarytrees_htmlwidget

I feel obligated to mention this again. If you want to get a preview of the exciting changes for htmlwidgets, be sure to check out htmlwidgets issue 86. Thanks to Joe Cheng from RStudio for taking the lead on this.

We can see some early examples with plotly from early adopter Andrew Clark @pssguy.

This Week’s Widget - summarytrees_htmlwidget

Kenneth Shirley and Howard Karloff authored an interesting paper.

(2013). Howard Karloff and Kenneth E. Shirley. “Maximum Entropy Summary Trees”, Computer Graphics Forum (Proc. EuroVis), Volume 32, Issue 3, Part 1, pp. 71-80.

Subsequently, Kenny made an R package summarytrees to create and visualize these summary trees. I learned about the R/d3.js combination after Kenny’s presentation at JSM 2015. Similar to the htmlwidget conversions of stmBrowser and stmCorrViz, I thought summarytrees would also be a great candidate for an htmlwidget.

This is a work in progress. Please offer feedback and ideas on this Github issue.

Installation

This is not on CRAN and only exists in a summarytrees fork, so to install we will need some help from devtools::install_github.

devtools::install_github("timelyportfolio/summarytrees@htmlwidget")

Examples

# devtools::install_github("timelyportfolio/summarytrees@htmlwidget")

library(summarytrees)
data(dmoz)

#use example from vignette
K <- 100
g <- greedy(node = dmoz[, "node"],
           parent = dmoz[, "parent"],
           weight = dmoz[, "weight"],
           label = dmoz[, "label"],
           K = K)

# Prepare the summary trees for visualization:
json <- prepare.vis(tree.list = g$summary.trees,
                    labels = g$data[, "label"],
                    tree = g$tree,
                    legend.width = 150,
                    node.width = 225,
                    node.height = 14,
                    units = "# of URLs",
                    print.weights = TRUE,
                    legend.color = "lightsteelblue",
                    color.level = 3)
summarytrees_htmlwidget(json)

## [1] "Running order.nodes() function to prepare data"
## [1] "  Computing node levels"
## [1] "  Computing child indices for each parent"
## [1] "Running C function to compute summary trees"
## [1] "Computation finished; now formatting output"

Thanks

Thanks Kenneth Shirley for the research and implementation of summarytrees.

As always, thanks to