shinyBN is an R/Shiny application for interactive inference and visualization of Bayesian network.

It's based on five R packages:

bnlearn: Structure learning and parameter training

gRain: Network inference

visNetwork: Network visualization

pROC: Receiver Operating Characteristic (ROC) curve plot

rmda: Decision Curves Analysis (DCA) plot

Which is further wrapped by `R/Shiny`. It's friendly to users and can easily use through the following workflow :

Select grade interval and color:

Yes?

Input the interval:

Input the color:

Notes: Separated by a comma.

Input the label:

Label Size:

Notes: Separated by a comma. If not, please input NULL.

Input the X position:

Input the Y position:

Notes:For x: Number from 0 (left) to 100 (right); For y: Number from 0 (bottom) to 100 (top).

W:

Smooth?

ROC Color:

ROC line type:

ROC width

Best Cutoff?

Cutoff Color:

Cutoff Size:

AUC?

CI of AUC?

AUC Color:

AUC Size:

Display the area as a polygon?

Polygon Color:

Add grid line?

Grid line Color:

Grid line type:

Grid line width

standardized net benefit?

DCA name:

Legend position:

Line Color:

Line Linetype:

Line Width:

Notes: Separated by a comma. The first element correspond to the curves provided. The next two elements correspond to the attributes of the 'all' and 'none' curves.

Label of x-axis:

Min, Max of x-axis:

Min, Max of y-axis:

shinyBN is an R/Shiny application for interactive inference and visualization of Bayesian network.

It's based on five R packages:

Which is further wrapped by `R/Shiny`. It's friendly to users and can easily use through the following workflow :

Select the type of Input:

Structure learning:

Parameter learning:

Input the interval:

Input the color:

Input the label:

W:

H:

W:

H:

shinyBN is an R/Shiny application for interactive inference and visualization of Bayesian network.

It's based on five R packages:

Which is further wrapped by R/Shiny. It's friendly to users and can easily use through the following workflow :

Select the type of Input:

Structure learning:

Parameter learning:

Input the interval:

Input the color:

Input the label:

W:

H:

W:

H:

Which is further wrapped by `R/Shiny`. It's friendly to users and can easily use through the following workflow :