In GICUNED/GridFCM: Fuzzy Cognitive Maps (FCM) for RepGrids and ImpGrids

knitr::opts_chunk$set(echo = TRUE)
library(GridFCM)
library(OpenRepGrid)
library(knitr)
library(rgl)
library(shiny)
library(plotly)
knitr::knit_hooks$set(webgl = hook_webgl)

GRIDS {data-orientation=rows}

sidebar {.sidebar}

GRIDFCM REPORT

Grids {.tabset .tabset-fade}

RepGrid

bertin(render.grid, colors = c("white","#005440"))

ImpGrid

bertin(render.imp, colors = c("#F52722","#e5e5e5","#a5d610"))

Biplot {data-navmenu="REPGRID ANALYSIS"}

sidebar {.sidebar}

BIPLOTS

2 {.tabset .tabset-fade}

Biplot Simple

biplotSimple(render.grid, c.label.col = "darkred")

Biplot 2D

biplot2d(render.grid, c.label.col = "darkred")

Biplot 3D

biplot3d(render.grid, c.label.col = "darkred")

Cluster Analysis {data-navmenu="REPGRID ANALYSIS"}

sidebar {.sidebar}

CLUSTER ANALYSIS

Constructos

Constructs

cluster(render.grid, along = 1)

Elementos

Elements

cluster(render.grid, along = 2)

Cognitive Indeces {data-navmenu="REPGRID ANALYSIS" data-orientation=rows}

sidebar {.sidebar}

COGNITIVE INDECES

Global {data-height=50}

Indeces

COR.IDEAL <- elementCor(render.grid)[1,ncol(render.grid)]
PVEFF <- indexPvaff(render.grid)
INTc <- indexIntensity(render.grid)[[1]]
INTc <- INTc[order(-INTc)]
INTe <- indexIntensity(render.grid)[[2]]
INTe <- INTe[order(-INTe)]
INT <- indexIntensity(render.grid)[[5]]
CON <- indexConflict1(render.grid)[[4]]

kable(data.frame(COR.IDEAL,PVEFF,INT,CON),col.names = c("Cor. Self / Ideal-Self","PVAFF","Intensity","Conflicts"))

Intensity {.tabset .tabset-fade}

Construct Intensity

kable(INTc, col.names = "Intensity")

Element Intensity

kable(INTe, col.names = "Intensity")

Implicative Dilemmas {data-navmenu="REPGRID ANALYSIS"}

dilemma <- indexDilemma(render.grid)

sidebar {.sidebar}

IMPLICATIVE DILEMMAS

Congruent/Discrepant Constructs

Congruent/Discrepant Constructs

colnames(dilemma$res1)[1] <- "Status"
kable(dilemma$res1)

Dilemmas

Dilemmas

kable(dilemma$res4)

Ideal Map Digraph {data-navmenu="FCM ANALYSIS"}

sidebar {.sidebar}

IDEAL MAP DIGRAPH

Ideal Map Digraph {.tabset .tabset-fade}

Full

idealdigraph(render.grid,render.imp)

Inconsistencies

idealdigraph(render.grid,render.imp, inc=TRUE)

Self Map Digraph {data-navmenu="FCM ANALYSIS" data-orientation=rows}

sidebar {.sidebar}

sliderInput('iter', 'Choose iteration:', min=1, max=30,
            value=0, step=1, round=0, animate=TRUE)

animationOptions(interval = 1000, loop = TRUE, playButton = TRUE,
  pauseButton = TRUE)

SELF MAP DIGRAPH

FCM Digraph {.tabset .tabset-fade}

Multidimensional Scaling 3D

rgl.clear()
fcmdigraph3D(render.grid,render.imp, niter = 1)

Multidimensional Scaling

fcmdigraph(render.grid,render.imp,niter = 1, layout = "mds")

Graphopt

renderPlot({
fcmdigraph(render.grid,render.imp,niter = input$iter, layout = "graphopt")
})  

Reingold - Tilford

fcmdigraph(render.grid,render.imp,niter = 1, layout = "rtcircle")

Circle

fcmdigraph(render.grid,render.imp,niter = 1, layout = "circle")

Grid

fcmdigraph(render.grid,render.imp,niter = 1, layout = "grid")

Personal Construct System Dynamics {data-navmenu="FCM ANALYSIS" data-orientation=rows}

sidebar {.sidebar}

PERSONAL CONSTRUCT SYSTEM DYNAMICS

PCSD Plot

PCSD plot

pcsd(render.grid,render.imp)

PCSD Indeces {.tabset .tabset-fade}

Summary

sum.pcsd <- pcsd_summary(render.grid, render.imp)
sum.pcsd <- sum.pcsd[order(-sum.pcsd[,3]),]

kable(sum.pcsd)

AUC

auc.pcsd <- auc_index(render.grid, render.imp)
auc.pcsd <- auc.pcsd[order(-auc.pcsd)]

kable(auc.pcsd, col.names = "AUC")

Stability

sd.pcsd <- stability_index(render.grid, render.imp)
sd.pcsd <- sd.pcsd[order(sd.pcsd)]

kable(sd.pcsd, col.names = "SD")

PCSD Derivative

pcsd_derivative(render.grid,render.imp)

FCM Indeces {data-navmenu="FCM ANALYSIS" data-orientation=rows}

sidebar {.sidebar}

FCM INDECES

1 {data-height=50}

ImpGrid Indeces

n.edges <- sum(getRatingLayer(render.imp) != 0)
n.vertex <- ncol(getRatingLayer(render.imp))
dens <- density_index(render.imp)
degree <- cbind(degree_index(render.imp)[[1]],degree_index(render.imp)[[2]],degree_index(render.imp)[[3]])
degree <- degree[order(-degree[,3]),]
close <- close_index(render.imp)
close <- close[order(-close)]
betw <- betw_index(render.imp)
betw <- betw[order(-betw)]

kable(data.frame(n.vertex,n.edges, dens),col.names = c("Vertexes","Edges","Density"))

2 {.tabset .tabset-fade}

Degree

kable(degree, col.names = c("Outputs","Inputs","All"))

Closeness

kable(close, col.names = "Centrality")

Betweenness

kable(betw, col.names = "Centrality")

REFERENCES

GICUNED/GridFCM documentation built on Feb. 23, 2023, 9:03 a.m.