Trying out new functionality; working on reparing traces

code/01_preprocessing_8o8m.R

@@ -20,8 +20,7 @@ datraw2 <- datraw[!is.na(datraw$artwork), ]
 # TODO: Why is this happening?
 
 # convert to log events
-datlogs <- create_eventlogs(datraw2, xmlfiles = paste0(artworks, "_de.xml"),
+datlogs <- create_eventlogs(datraw2, xmlpath = "../data/8o8m/Content8o8m/")
-                            xmlpath = "../data/8o8m/Content8o8m/")
 
 # export data
 write.table(datlogs, paste0("../data/8o8m/event_logfiles_", now, ".csv"),

code/01_preprocessing_haum.R

@@ -6,25 +6,25 @@ devtools::load_all("../../../../software/mtt")
 now <- format(Sys.time(), "%Y-%m-%d_%H-%M-%S")
 #now <- "2023-09-23_01-31-30"
 
-path <- "../data/haum/LogFiles/"
+#--------------- Parse raw log files ---------------
 
+path <- "../data/haum/LogFiles/"
 folders <- dir(path)
 
-# parse raw log files
 #datraw <- parse_logfiles(folders, path)
 datraw <- read.table("../data/haum/raw_logfiles_small_2023-09-26_13-50-20.csv",
                      sep = ";", header = TRUE)
 
 # export data
-write.table(datraw, paste0("../data/haum/raw_logfiles_", now, ".csv"),
+# write.table(datraw, paste0("../data/haum/raw_logfiles_", now, ".csv"),
-            sep = ";", row.names = FALSE)
+#             sep = ";", row.names = FALSE)
 
-# convert to log events
+#--------------- Create event logs ---------------
-datlogs <- create_eventlogs(datraw, 
+
+datlogs <- create_eventlogs(datraw,
            xmlpath = "../data/haum/ContentEyevisit/eyevisit_cards_light/")
 
 artworks <- unique(na.omit(datlogs$artwork))
-artworks <- artworks[!artworks %in% c("504", "505")]
 topics <- extract_topics(artworks, xmlfiles = paste0(artworks, ".xml"),
                          xmlpath = "../data/haum/ContentEyevisit/eyevisit_cards_light/")
 
@@ -32,6 +32,72 @@ datlogs_topics <- add_topic(datlogs, topics = topics)
 
 
 # export data
-write.table(datlogs_topics, paste0("../data/haum/event_logfiles_", now, ".csv"),
+# write.table(datlogs_topics, paste0("../data/haum/event_logfiles_", now, ".csv"),
+#             sep = ";", row.names = FALSE)
+
+#--------------- Add meta data ---------------
+
+## Read data for holiday
+
+hd0 <- read.table("../data/metadata/feiertage.csv", sep = ";", header = TRUE)
+hd0$X.br. <- NULL
+
+hd <- hd0[hd0$Abkuerzung == "NI", ]
+names(hd) <- c("state", "stateCode", "date", "holiday")
+hd$date <- as.POSIXct(hd$date)
+
+## Read data for school vacations
+
+# https://ferien-api.de/#holidaysPerStateAndYear
+# Data extracted (on Linux) via:
+# curl https://ferien-api.de/api/v1/holidays/NI > schulferien_NI.json
+
+# library(jsonlite)
+#
+# dat <- read_json("data/metadata/schulferien_NI.json", simplify = TRUE)
+# dat$slug <- NULL
+#
+# dat$name <- paste0(gsub("^(.*).niedersachsen.*", "\\1", dat$name),
+#                    gsub("^.*niedersachsen [0-9]{4}(.*)", "\\1",
+#                               dat$name))
+#
+# write.table(dat, "data/metadata/schulferien_2019-2025_NI.csv", sep = ";",
+#             row.names = FALSE, quote = FALSE)
+
+sf1 <- read.table("../data/metadata/schulferien_2016-2018_NI.csv", sep = ";",
+                  header = TRUE)
+sf2 <- read.table("../data/metadata/schulferien_2019-2025_NI.csv", sep = ";",
+                  header = TRUE)
+sf <- rbind(sf1, sf2)
+sf$start <- as.Date(sf$start)
+sf$end <- as.Date(sf$end)
+
+sfdat <- NULL
+
+for (i in seq_len(nrow(sf))) {
+  date <- seq(sf$start[i], sf$end[i], by = 1)
+  sfdat <- rbind(sfdat, data.frame(date, vacations = sf$name[i],
+                                   stateCodeVacations = sf$stateCode[i]))
+}
+
+# TODO: How to handle stateCode? There will be several for certain types of
+# data sets... Not important here, since I only do NI.
+
+## Add metadata
+
+# holidays
+dat1 <- merge(datlogs_topics, hd, by.x = "date.start", by.y = "date", all.x = TRUE)
+# school vacations
+dat2 <- merge(dat1, sfdat, by.x = "date.start", by.y = "date", all.x = TRUE)
+
+## Export data
+
+write.table(dat2,
+            file = paste0("../data/haum/event_logfiles_small_metadata_", now, ".csv"),
             sep = ";", row.names = FALSE)
 
+# TODO: Maybe add infos about artworks?
+
+
+
+

code/02_metadata_haum.R

@@ -48,7 +48,7 @@ for (i in seq_len(nrow(sf))) {
 
 # load (small) event log data set
 #dat <- read.table("data/haum/event_logfiles_2023-09-23_01-31-30.csv",
-dat <- read.table("data/haum/event_logfiles_small_2023-10-15_10-08-43.csv",
+dat <- read.table("data/haum/event_logfiles_small_2023-10-19_18-25-26.csv",
                   sep = ";", header = TRUE)
 dat$date.start <- as.POSIXct(dat$date.start)
 dat$date.stop <- as.POSIXct(dat$date.stop)
@@ -65,7 +65,7 @@ dat2 <- merge(dat1, sfdat, by = "date", all.x = TRUE)
 ## Export data
 
 write.table(dat2,
-            file = "data/haum/event_logfiles_small_metadata_2023-10-15_10-08-43.csv",
+            file = "data/haum/event_logfiles_small_metadata_2023-10-19_18-25-26.csv",
             sep = ";", row.names = FALSE)
 
 # TODO: Maybe add infos about artworks?

code/04_modeling_haum.R

@@ -4,7 +4,7 @@
 
 # dat0 <- read.table("../data/haum/event_logfiles_metadata_2023-09-23_01-31-30.csv",
 #                   sep = ";", header = TRUE)
-dat0 <- read.table("../data/haum/event_logfiles_small_metadata_2023-10-15_10-08-43.csv",
+dat0 <- read.table("../data/haum/event_logfiles_small_metadata_2023-10-19_18-25-26.csv",
                   sep = ";", header = TRUE)
 dat0$date <- as.Date(dat0$date)
 dat0$date.start <- as.POSIXct(dat0$date.start)
@@ -81,7 +81,7 @@ datrm[datrm$trace %in% ct, 1:10]
 
 ### WHY?????
 
-alog <- activitylog(datrm,
+alog <- activitylog(dat,
                     case_id = "trace",
                     activity_id = "event",
                     #resource_id = "case",