Visualising and Analysing Geographic and Movement Data
In HW5 we will mainly focus on answering the bullet point 1 and 2 of challenge 2 in VAST 2022, ie.
Loading packages
packages = c('tidyverse','knitr','tmap','readr','sf','clock')
for(p in packages){
if(!require(p, character.only = T)){
install.packages(p)
}
library(p, character.only = T)
}
Data reading with read_sf function.
schools <- read_sf("data/Attributes/Schools.csv",
options = "GEOM_POSSIBLE_NAMES=location")
pubs <- read_sf("data/Attributes/Pubs.csv",
options = "GEOM_POSSIBLE_NAMES=location")
apartments <- read_sf("data/Attributes/Apartments.csv",
options = "GEOM_POSSIBLE_NAMES=location")
buildings <- read_sf("data/Attributes/Buildings.csv",
options = "GEOM_POSSIBLE_NAMES=location")
employers <- read_sf("data/Attributes/Employers.csv",
options = "GEOM_POSSIBLE_NAMES=location")
restaurants <- read_sf("data/Attributes/Restaurants.csv",
options = "GEOM_POSSIBLE_NAMES=location")
Building layout by basic funtions
tmap_mode("view")
tm_shape(buildings)+
tm_polygons(col = "buildingType",
palette="Accent",
border.col = "black",
border.alpha = .5,
border.lwd = 0.5)
tmap_mode("view")
Apartments layout by Rental Cost.
apartments$rentalCost <- as.numeric(apartments$rentalCost)
apartments$numberOfRooms <- as.numeric(apartments$numberOfRooms)
apartments$maxOccupancy <- as.numeric(apartments$maxOccupancy)
tmap_mode("view")
tm_shape(buildings)+
tm_polygons(col = "buildingType",
palette="Accent",
border.col = "black",
border.alpha = 0.4,
border.lwd = 0.3)+
tm_shape(apartments)+
tm_bubbles(col = "rentalCost",
alpha = 0.4,
n = 4,
style = "jenks",
palette="OrRd",
size = "numberOfRooms",
scale = 0.4,
border.col = "black",
border.lwd = 0.3)
tmap_mode("view")
Employers layout by # of ppl employed (headcouts).
Jobs<-readRDS("data/Jobs.rds")
employers$employerId <- as.numeric(employers$employerId)
employers$buildingId <- as.numeric(employers$buildingId)
weeklypay_employer <- Jobs %>%
group_by(employerId) %>%
summarise(
n=n(),
mean=mean(totalweeklypay))
colnames(weeklypay_employer)[2] <- "headcount"
colnames(weeklypay_employer)[3] <- "averagepay"
employers<-weeklypay_employer %>%
select("employerId","headcount","averagepay") %>%
distinct() %>%
left_join(employers, by = 'employerId')
employers1<-st_as_sf(employers)
tmap_mode("view")
tm_shape(buildings)+
tm_polygons(col = "buildingType",
palette="Accent",
border.col = "black",
border.alpha = 0.4,
border.lwd = 0.3)+
tm_shape(employers1)+
tm_bubbles(col = "headcount",
alpha = 0.4,
n = 4,
style = "cont",
palette="OrRd",
scale = 0.4,
border.col = "black",
border.lwd = 0.3)
tmap_mode("view")
Employers layout by weekly average salary.
tmap_mode("view")
tm_shape(buildings)+
tm_polygons(col = "buildingType",
palette="Accent",
border.col = "black",
border.alpha = 0.4,
border.lwd = 0.3)+
tm_shape(employers1)+
tm_bubbles(col = "averagepay",
alpha = 0.4,
n = 4,
style = "cont",
palette="viridis",
scale = 0.4,
border.col = "black",
border.lwd = 0.3)
tmap_mode("view")
Pubs layout by hourly cost.
pubs$hourlyCost <- as.numeric(pubs$hourlyCost)
tmap_mode("view")
tm_shape(buildings)+
tm_polygons(col = "buildingType",
palette="Accent",
border.col = "black",
border.alpha = 0.4,
border.lwd = 0.3)+
tm_shape(pubs)+
tm_bubbles(col = "hourlyCost",
alpha = 0.4,
n = 4,
style = "jenks",
palette="Reds",
scale = 0.4,
border.col = "black",
border.lwd = 0.3)
tmap_mode("view")
Schools layout by monthly cost.
schools$monthlyCost <- as.numeric(schools$monthlyCost)
tmap_mode("view")
tm_shape(buildings)+
tm_polygons(col = "buildingType",
palette="Accent",
border.col = "black",
border.alpha = 0.4,
border.lwd = 0.3)+
tm_shape(schools)+
tm_bubbles(col = "monthlyCost",
alpha = 0.4,
n = 4,
style = "jenks",
palette="Blues",
scale = 0.4,
border.col = "black",
border.lwd = 0.3)
tmap_mode("view")
From travelJournal we can find out the frequency of building ID’s occurrence. We have cleaned the data to the minimum size that includes the location geometric information. However, the hex method failed to capture the occurrence of the geometric ID, most of the occurrence fell in NA category. Thus below table is a gimps of a clean data table that includes occurrence and location ID.
travellocation<-readRDS("data/travellocation2.rds")
knitr::kable(head(travellocation), format = 'html')
| travelStartLocationId | location |
|---|---|
| 1 | POLYGON ((350.0639 4595.666… |
| 1 | POLYGON ((350.0639 4595.666… |
| 1 | POLYGON ((350.0639 4595.666… |
| 1 | POLYGON ((350.0639 4595.666… |
| 1 | POLYGON ((350.0639 4595.666… |
| 1 | POLYGON ((350.0639 4595.666… |