Cleaned up preprocessing file for HAUM
This commit is contained in:
parent
cd163040d6
commit
4973ceb31b
|
|
@ -1,3 +1,19 @@
|
|||
# 01_preprocessing_haum.R
|
||||
#
|
||||
# content: (1) Parse raw log files
|
||||
# (2) Create event logs
|
||||
# (3) Add meta data
|
||||
#
|
||||
# input: raw log files from ../data/haum/*.log
|
||||
# ../data/metadata/feiertage.csv
|
||||
# ../data/metadata/schulferien_2016-2018_NI.csv
|
||||
# ../data/metadata/schulferien_2019-2025_NI.csv
|
||||
# output: raw_logfiles_<timestamp>.csv
|
||||
# event_logfiles_<timestamp>.csv
|
||||
# event_logfiles_<timestamp>.csv
|
||||
#
|
||||
# last mod: 2023-10-23, NW
|
||||
|
||||
# setwd("C:/Users/nwickelmaier/Nextcloud/Documents/MDS/2023ss/60100_master_thesis/code")
|
||||
|
||||
#library(mtt)
|
||||
|
|
@ -6,20 +22,19 @@ devtools::load_all("../../../../software/mtt")
|
|||
now <- format(Sys.time(), "%Y-%m-%d_%H-%M-%S")
|
||||
#now <- "2023-09-23_01-31-30"
|
||||
|
||||
#--------------- Parse raw log files ---------------
|
||||
#--------------- (1) Parse raw log files ---------------
|
||||
|
||||
path <- "../data/haum/LogFiles/"
|
||||
folders <- dir(path)
|
||||
|
||||
#datraw <- parse_logfiles(folders, path)
|
||||
datraw <- read.table("../data/haum/raw_logfiles_small_2023-09-26_13-50-20.csv",
|
||||
sep = ";", header = TRUE)
|
||||
datraw <- parse_logfiles(folders, path)
|
||||
|
||||
# export data
|
||||
# write.table(datraw, paste0("../data/haum/raw_logfiles_", now, ".csv"),
|
||||
# sep = ";", row.names = FALSE)
|
||||
## Export data
|
||||
|
||||
#--------------- Create event logs ---------------
|
||||
write.table(datraw, paste0("../data/haum/raw_logfiles_", now, ".csv"),
|
||||
sep = ";", row.names = FALSE)
|
||||
|
||||
#--------------- (2) Create event logs ---------------
|
||||
|
||||
datlogs <- create_eventlogs(datraw,
|
||||
xmlpath = "../data/haum/ContentEyevisit/eyevisit_cards_light/")
|
||||
|
|
@ -30,12 +45,7 @@ topics <- extract_topics(artworks, xmlfiles = paste0(artworks, ".xml"),
|
|||
|
||||
datlogs_topics <- add_topic(datlogs, topics = topics)
|
||||
|
||||
|
||||
# export data
|
||||
# write.table(datlogs_topics, paste0("../data/haum/event_logfiles_", now, ".csv"),
|
||||
# sep = ";", row.names = FALSE)
|
||||
|
||||
#--------------- Add meta data ---------------
|
||||
#--------------- (3) Add meta data ---------------
|
||||
|
||||
## Read data for holiday
|
||||
|
||||
|
|
@ -83,21 +93,15 @@ for (i in seq_len(nrow(sf))) {
|
|||
# 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
|
||||
## Merge data
|
||||
|
||||
# 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"),
|
||||
write.table(dat2, paste0("../data/haum/event_logfiles_", now, ".csv"),
|
||||
sep = ";", row.names = FALSE)
|
||||
|
||||
# TODO: Maybe add infos about artworks?
|
||||
|
||||
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -1,72 +0,0 @@
|
|||
# setwd("C:/Users/nwickelmaier/Nextcloud/Documents/MDS/2023ss/60100_master_thesis/")
|
||||
|
||||
## 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.
|
||||
|
||||
# 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_2023-10-19_18-25-26.csv",
|
||||
sep = ";", header = TRUE)
|
||||
dat$date.start <- as.POSIXct(dat$date.start)
|
||||
dat$date.stop <- as.POSIXct(dat$date.stop)
|
||||
dat$artwork <- sprintf("%03d", dat$artwork)
|
||||
dat$date <- as.Date(dat$date.start)
|
||||
|
||||
## Add metadata
|
||||
|
||||
# holidays
|
||||
dat1 <- merge(dat, hd, by = "date", all.x = TRUE)
|
||||
# school vacations
|
||||
dat2 <- merge(dat1, sfdat, by = "date", all.x = TRUE)
|
||||
|
||||
## Export data
|
||||
|
||||
write.table(dat2,
|
||||
file = "data/haum/event_logfiles_small_metadata_2023-10-19_18-25-26.csv",
|
||||
sep = ";", row.names = FALSE)
|
||||
|
||||
# TODO: Maybe add infos about artworks?
|
||||
|
||||
|
|
@ -33,5 +33,5 @@ points(y.stop ~ x.stop, datmove, xlim = c(0, 3840), ylim = c(0, 2160), col = "bl
|
|||
points(ty ~ tx, dat, xlim = c(0, 3840), ylim = c(0, 2160), col = "red")
|
||||
legend("topleft", c("start", "stop", "trans"), col = c("black", "blue", "red"), pch = 21)
|
||||
|
||||
# --> does not seem to be a good way to compare if trans values ae valid
|
||||
# --> does not seem to be a good way to compare *if* trans values are valid
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue