Added some plots etc.

This commit is contained in:
Nora Wickelmaier 2024-03-14 17:28:10 +01:00
parent 752501b4df
commit 88ac019b93

View File

@ -8,7 +8,7 @@
# input: results/haum/event_logfiles_2024-02-21_16-07-33.csv # input: results/haum/event_logfiles_2024-02-21_16-07-33.csv
# output: results/haum/eventlogs_pre-corona_case-clusters.csv # output: results/haum/eventlogs_pre-corona_case-clusters.csv
# #
# last mod: 2024-03-08 # last mod: 2024-03-14
# setwd("C:/Users/nwickelmaier/Nextcloud/Documents/MDS/2023ss/60100_master_thesis/analysis/code") # setwd("C:/Users/nwickelmaier/Nextcloud/Documents/MDS/2023ss/60100_master_thesis/analysis/code")
@ -52,17 +52,19 @@ hc <- hcs$ward
k <- 5 k <- 5
mycols <- c("#434F4F", "#78004B", "#FF6900", "#3CB4DC", "#91C86E") mycols <- c("#3CB4DC", "#FF6900", "#78004B", "#91C86E", "#434F4F")
cluster <- cutree(as.hclust(hc), k = k) cluster <- cutree(as.hclust(hc), k = k)
table(cluster) table(cluster)
plot(coor_2d, col = mycols[cluster]) plot(coor_2d, col = mycols[cluster], pch = 16)
legend("topleft", paste("Cl", 1:4), col = mycols, pch = 21) #legend("topleft", paste("Cl", 1:5), col = mycols, pch = 21)
legend("topleft", c("Scanning", "Exploring", "Flitting", "Searching", "Info"),
col = mycols, bty = "n", pch = 16)
rgl::plot3d(coor_3d, col = mycols[cluster]) rgl::plot3d(coor_3d, col = mycols[cluster])
ftable(xtabs( ~ InfocardOnly + Pattern + cluster, dattree)) print(ftable(xtabs( ~ InfocardOnly + Pattern + cluster, dattree)), zero = "-")
aggregate(. ~ cluster, df, mean) aggregate(. ~ cluster, df, mean)
@ -124,6 +126,10 @@ write.table(res,
save(res, dist_mat, hcs, acs, coor_2d, coor_3d, save(res, dist_mat, hcs, acs, coor_2d, coor_3d,
file = "results/haum/tmp_user-navigation.RData") file = "results/haum/tmp_user-navigation.RData")
save(coor_2d, coor_3d, cluster,
file = "../../thesis/figures/data/clustering_cases.RData")
#--------------- (3) Fit tree --------------- #--------------- (3) Fit tree ---------------
c1 <- rpart::rpart(as.factor(cluster) ~ ., data = dattree[, c("PropMoves", c1 <- rpart::rpart(as.factor(cluster) ~ ., data = dattree[, c("PropMoves",
@ -134,8 +140,8 @@ c1 <- rpart::rpart(as.factor(cluster) ~ ., data = dattree[, c("PropMoves",
"InfocardOnly")], "InfocardOnly")],
method = "class") method = "class")
pdf("results/figures/tree_items_rpart.pdf", height = 5, width = 15, pointsize = 10) pdf("results/figures/tree_cases_rpart.pdf", height = 5, width = 15, pointsize = 10)
plot(partykit::as.party(c1)) plot(partykit::as.party(c1), tp_args = list(fill = mycols, col = mycols))
dev.off() dev.off()
# with conditional tree # with conditional tree
@ -147,7 +153,20 @@ c2 <- partykit::ctree(as.factor(cluster) ~ ., data = dattree[, c("PropMoves",
"InfocardOnly")], "InfocardOnly")],
alpha = 0.001) alpha = 0.001)
pdf("results/figures/tree_items_ctree.pdf", height = 7, width = 20, pointsize = 10) pdf("results/figures/tree_cases_ctree.pdf", height = 7, width = 20, pointsize = 10)
plot(c2) plot(c2, tp_args = list(fill = mycols, col = mycols))
dev.off() dev.off()
factoextra::fviz_dend(as.hclust(hc), k = k,
cex = 0.5,
k_colors = mycols,
#type = "phylogenic",
rect = TRUE,
main = "",
ylab = ""
#ggtheme = ggplot2::theme_bw()
)