Merge pull request #113 from Hong-Kong-Districts-Info/feat/21-data-up…

…date

Feat: Add 2020 and 2021 collision data

inst/app/global.R

@@ -57,7 +57,7 @@ i18n$use_js()
 # Data import -------------------------------------------------------------
 
 ## Take data from {hkdatasets}
-hk_accidents <- fst::read_fst("./data/hk_accidents.fst")
+hk_collisions <- fst::read_fst("./data/hk_collisions.fst")
 hk_vehicles <- fst::read_fst("./data/hk_vehicles.fst")
 hk_casualties <- fst::read_fst("./data/hk_casualties.fst")
 
@@ -67,7 +67,7 @@ hotzone_streets = read_sf("./data/hotzone_streets.gpkg")
 terminology = read.csv("./data/terminology.csv")
 
 ## Manipulated data, generated from `modules/manipulate_data.R`
-hk_accidents_valid_sf = read_sf("./data/data-manipulated/hk_accidents_valid_sf.gpkg")
+hk_collisions_valid_sf = read_sf("./data/data-manipulated/hk_collisions_valid_sf.gpkg")
 hotzone_out_df = fst::read_fst("./data/data-manipulated/hotzone_out_df.fst")
 
 # interactive thematic map mode option ------------------------------------
@@ -86,7 +86,7 @@ DISTRICT_FULL_NAME = hkdatasets::hkdistrict_summary[["District_EN"]]
 # Color scheme ------------------------------------------------------------
 
 SEVERITY_COLOR = c(Fatal = "#FF4039", Serious = "#FFB43F", Slight = "#FFE91D")
-CATEGORY_COLOR = setNames(as.list(c("#232323", "#232323", "#232323")), c("accidents", "casualties", "vehicles"))
+CATEGORY_COLOR = setNames(as.list(c("#232323", "#232323", "#232323")), c("collisions", "casualties", "vehicles"))
 
 # Fill color palette according to the severity of the accident
 fill_palette = leaflet::colorFactor(palette = c("#FF4039", "#FFB43F", "#FFE91D"), domain = c("Fatal", "Serious", "Slight"))

inst/app/modules/district_dsb.R

@@ -3,45 +3,46 @@
 # Translation terms
 
 COLLISION_SEVERITY_TRANSLATE = data.frame(
-  Severity = c("Slight", "Serious", "Fatal"),
-  Severity_chi = c("輕微", "嚴重", "致命")
+  severity = c("Slight", "Serious", "Fatal"),
+  severity_chi = c("輕微", "嚴重", "致命")
 )
 
 COLLISION_TYPE_TRANSLATE = data.frame(
-  Collision_Type = c(
+  collision_type_with_cycle = c(
     "Vehicle collision with Vehicle", "Vehicle collision with Pedestrian", "Vehicle collision with Pedal Cycle",
     "Vehicle collision with Object", "Vehicle collision with Nothing", "Pedal Cycle collision with Pedestrian",
-    "Pedal Cycle collision with Pedal Cycle", "Pedal Cycle collision with Object", "Pedal Cycle collision with Nothing"
+    "Pedal Cycle collision with Pedal Cycle", "Pedal Cycle collision with Object", "Pedal Cycle collision with Nothing",
+    "Unknown vehicle collision type"
   ),
-  Collision_Type_chi = c(
+  collision_type_with_cycle_chi = c(
     "車撞車", "車撞行人", "車撞單車",
     "車撞物", "車輛沒有碰撞", "單車撞行人",
-    "單車撞單車", "單車撞物", "單車沒有碰撞"
+    "單車撞單車", "單車撞物", "單車沒有碰撞",
+    "類別不明"
   )
 )
 
 VEHICLE_CLASS_TRANSLATE = data.frame(
-  Vehicle_Class = c(
+  vehicle_class = c(
     "Private car", "Public franchised bus", "Taxi", "Motorcycle", "Light goods vehicle",
     "Bicycle", "Heavy goods vehicle", "Medium goods vehicle", "Tram", "Public light bus",
     "Others (incl. unknown)", "Public non-franchised bus", "Light rail vehicle"),
-  Vehicle_Class_chi = c(
+  vehicle_class_chi = c(
     "私家車", "公共專營巴士", "的士", "電單車", "輕型貨車",
     "單車", "重型貨車", "中型貨車", "電車", "公共小巴",
     "其他（包括類別不詳車輛）", "公共非專營巴士", "輕鐵車輛"
   )
 )
 
-# Unique Main_vehicle values extracted from hk_vehicles
-# TODO: Unify the capitalisation rules
+# Unique vehicle movement values extracted from hk_vehicles
 VEHICLE_MOVEMENT_TRANSLATE = data.frame(
-  Main_vehicle = c(
+  vehicle_movement = c(
     "Going straight ahead (with priority)", "Changing lanes or merging", "Overtaking on off-side",
     "Overtaking on near-side", "Going Straight Ahead (against priority)", "Making right turn", "Making left turn",
     "Making U turn", "Slowing or stopping", "Stopped in traffic", "Starting in traffic", "Leaving parking place",
     "Parked", "Reversing", "Driverless moving vehicle", "Ran off road", "Other", "Unknown"
   ),
-  Main_vehicle_chi = c(
+  vehicle_movement_chi = c(
     "向前駛 （優先）", "轉換行車線", "從外線超車",
     "從內線超車", "向前駛（無優先）", "右轉", "左轉",
     "掉頭", "慢駛或停車", "因前路受阻而停車", "跟隨前面交通開車", "駛離泊車位",
@@ -50,21 +51,21 @@ VEHICLE_MOVEMENT_TRANSLATE = data.frame(
 )
 
 PED_ACTION_TRANSLATE = data.frame(
-  Ped_Action = c(
+  ped_action = c(
     "Walking - back to traffic", "Walking - facing traffic", "Standing", "Boarding vehicle",
     "Alighting from vehicle", "Falling or jumping from vehicle", "Working at a vehicle", "Other working",
     "Playing", "Crossing from near-side", "Crossing from off-side", "Not known"
   ),
-  Ped_Action_chi = c(
+  ped_action_chi = c(
     "步行 ─ 背向車流", "步行 ─ 面向車流", "站立", "正在登車",
     "正在下車", "從車上跌下或跳下", "在修車中", "其他工作",
     "在玩耍中", "從車左邊橫過馬路", "從車右邊橫過馬路", "資料不詳"
   )
 )
 
 ROAD_HIERARCHY_TRANSLATE = data.frame(
-  Road_Hierarchy = c("Expressway", "Main Road", "Secondary Road", "Other Minor Road", "Cycle Track/Others"),
-  Road_Hierarchy_chi = c("快速公路", "主要道路", "次要道路（內街／支路）", "小路（鄉村道路／行人徑）", "單車徑/其他")
+  road_hierarchy = c("Expressway", "Main Road", "Secondary Road", "Other Minor Road", "Cycle Track/Others"),
+  road_hierarchy_chi = c("快速公路", "主要道路", "次要道路（內街／支路）", "小路（鄉村道路／行人徑）", "單車徑/其他")
 )
 
 
@@ -93,31 +94,31 @@ output$ksi_filter_ui = renderUI({
 })
 
 
-# Return filtered hk_accidents dataframe according to users' selected inputs
-ddsb_filtered_hk_accidents = reactive({
+# Return filtered hk_collisions dataframe according to users' selected inputs
+ddsb_filtered_hk_collisions = reactive({
   # filter by users' selected district
   # FIXME: Temp workaround to fix non-initialised value when district filter renders in server side
   ddsb_district_filter = if (is.null(input$ddsb_district_filter)) "CW" else input$ddsb_district_filter
-  hk_accidents_filtered = filter(hk_accidents, District_Council_District == ddsb_district_filter)
+  hk_collisions_filtered = filter(hk_collisions, district == ddsb_district_filter)
 
   # filter by users' selected time range
-  hk_accidents_filtered = filter(hk_accidents_filtered, Year >= input$ddsb_year_filter[1] & Year <= input$ddsb_year_filter[2])
+  hk_collisions_filtered = filter(hk_collisions_filtered, year >= input$ddsb_year_filter[1] & year <= input$ddsb_year_filter[2])
 
   # remove slightly injured collisions if user select "KSI only" option
   # FIXME: Temp workaround to fix non-initialised value when KSI filter renders in server side
   ddsb_ksi_filter = if (is.null(input$ddsb_ksi_filter)) "all" else input$ddsb_ksi_filter
 
   if (ddsb_ksi_filter == "ksi_only") {
-    hk_accidents_filtered = filter(hk_accidents_filtered, Severity != "Slight")
+    hk_collisions_filtered = filter(hk_collisions_filtered, severity != "Slight")
   }
 
   # Show only collisions with valid lng/lat
-  hk_accidents_filtered = hk_accidents_filtered %>%
-    filter(!is.na(Grid_E) & !is.na(Grid_N)) %>%
-    st_as_sf(coords = c("Grid_E", "Grid_N"), crs = 2326, remove = FALSE)
+  hk_collisions_filtered = hk_collisions_filtered %>%
+    filter(!is.na(easting) & !is.na(northing)) %>%
+    st_as_sf(coords = c("easting", "northing"), crs = 2326, remove = FALSE)
 
-  print(nrow(hk_accidents_filtered))
+  print(paste0("Number of records filtered in districh dashboard: ", nrow(hk_collisions_filtered)))
 
-  hk_accidents_filtered
+  hk_collisions_filtered
 
 })

inst/app/modules/district_dsb_all.R

@@ -5,29 +5,29 @@
 ddsb_filtered_hk_casualties = reactive({
 
   # vector of Serial No. in selected range
-  serial_no_filtered = unique(ddsb_filtered_hk_accidents()[["Serial_No_"]])
+  serial_no_filtered = unique(ddsb_filtered_hk_collisions()[["serial_no"]])
 
-  filter(hk_casualties, Serial_No_ %in% serial_no_filtered)
+  filter(hk_casualties, serial_no %in% serial_no_filtered)
 })
 
 # filtered hk_vehicles
 ddsb_filtered_hk_vehicles = reactive({
 
   # vector of Serial No. in selected range
-  serial_no_filtered = unique(ddsb_filtered_hk_accidents()[["Serial_No_"]])
+  serial_no_filtered = unique(ddsb_filtered_hk_collisions()[["serial_no"]])
 
-  filter(hk_vehicles, Serial_No_ %in% serial_no_filtered)
+  filter(hk_vehicles, serial_no %in% serial_no_filtered)
 })
 
 all_grid_count = reactive({
-  count_collisions_in_grid(ddsb_filtered_hk_accidents())
+  count_collisions_in_grid(ddsb_filtered_hk_collisions())
 })
 
 
 # Outputs ----------------------------------
 
 output$box_all_total_collision = renderInfoBox({
-  n_collision = nrow(ddsb_filtered_hk_accidents())
+  n_collision = nrow(ddsb_filtered_hk_collisions())
 
   infoBox(
     title = "",
@@ -51,7 +51,7 @@ output$box_all_total_casualty = renderInfoBox({
 })
 
 output$box_all_serious_stat = renderInfoBox({
-  n_serious = nrow(filter(ddsb_filtered_hk_casualties(), Degree_of_Injury == "Seriously Injured"))
+  n_serious = nrow(filter(ddsb_filtered_hk_casualties(), injury_degree == "Seriously Injured"))
   serious_per = round(n_serious / nrow(ddsb_filtered_hk_casualties()) * 100, digits = 1)
 
   infoBox(
@@ -64,7 +64,7 @@ output$box_all_serious_stat = renderInfoBox({
 })
 
 output$box_all_fatal_stat = renderInfoBox({
-  n_fatal = nrow(filter(ddsb_filtered_hk_casualties(), Degree_of_Injury == "Killed"))
+  n_fatal = nrow(filter(ddsb_filtered_hk_casualties(), injury_degree == "Killed"))
   fatal_per = round(n_fatal / nrow(ddsb_filtered_hk_casualties()) * 100, digits = 1)
 
   infoBox(
@@ -101,13 +101,13 @@ output$ddsb_all_collision_heatmap = renderTmap({
 output$ddsb_all_ksi_plot = renderPlotly({
 
   # count by severity
-  plot_data = count(ddsb_filtered_hk_accidents(), Severity, name = "count", na.rm = TRUE) %>%
-    left_join(COLLISION_SEVERITY_TRANSLATE, by = "Severity") %>%
+  plot_data = count(ddsb_filtered_hk_collisions(), severity, name = "count", na.rm = TRUE) %>%
+    left_join(COLLISION_SEVERITY_TRANSLATE, by = "severity") %>%
     # Force order of the categorical axis
     # Factor in reversed order since last element in factor is plotted on top in ggplot
-    mutate(Severity_text = factor(paste0(Severity_chi, "\n", Severity), c("致命\nFatal", "嚴重\nSerious", "輕微\nSlight")))
+    mutate(severity_text = factor(paste0(severity_chi, "\n", severity), c("致命\nFatal", "嚴重\nSerious", "輕微\nSlight")))
 
-  plot_by_severity = ggplot(plot_data, aes(x = Severity_text, y = count, fill = Severity)) +
+  plot_by_severity = ggplot(plot_data, aes(x = severity_text, y = count, fill = severity)) +
     geom_bar(stat = "identity") +
     coord_flip() +
     scale_fill_manual(values = SEVERITY_COLOR) +
@@ -125,9 +125,9 @@ output$ddsb_all_ksi_plot = renderPlotly({
 # Collision by year plot
 output$ddsb_all_year_plot = renderPlotly({
 
-  plot_data = count(ddsb_filtered_hk_accidents(), Year, name = "count", na.rm = TRUE)
+  plot_data = count(ddsb_filtered_hk_collisions(), year, name = "count", na.rm = TRUE)
 
-  collision_year_trend_plot = ggplot(plot_data, aes(x = Year, y = count)) +
+  collision_year_trend_plot = ggplot(plot_data, aes(x = year, y = count)) +
     geom_line() +
     theme_minimal() +
     theme(
@@ -144,15 +144,15 @@ output$ddsb_all_year_plot = renderPlotly({
 output$ddsb_all_collision_type_plot = renderPlotly({
 
   # count by pedestrian Action
-  plot_data = ddsb_filtered_hk_accidents() %>%
-    count(Type_of_Collision_with_cycle, name = "count") %>%
-    left_join(COLLISION_TYPE_TRANSLATE, by = c("Type_of_Collision_with_cycle" = "Collision_Type")) %>%
+  plot_data = ddsb_filtered_hk_collisions() %>%
+    count(collision_type_with_cycle, name = "count") %>%
+    left_join(COLLISION_TYPE_TRANSLATE, by = c("collision_type_with_cycle" = "collision_type_with_cycle")) %>%
     # Merge both en and zh values, then reorder vehicle class in descending order
-    mutate(Collision_Type_order = reorder(paste0(Collision_Type_chi, "\n", Type_of_Collision_with_cycle), count))
+    mutate(collision_type_with_cycle_order = reorder(paste0(collision_type_with_cycle_chi, "\n", collision_type_with_cycle), count))
 
 
-  plot_by_collision_type = ggplot(plot_data, aes(x = Collision_Type_order, y = count)) +
-    geom_bar(stat = "identity", fill = CATEGORY_COLOR$accidents) +
+  plot_by_collision_type = ggplot(plot_data, aes(x = collision_type_with_cycle_order, y = count)) +
+    geom_bar(stat = "identity", fill = CATEGORY_COLOR$collisions) +
     coord_flip() +
     theme_minimal() +
     theme(
@@ -170,12 +170,12 @@ output$ddsb_all_collision_type_plot = renderPlotly({
 output$ddsb_all_vehicle_class_plot = renderPlotly({
 
   # count by Vehicle_Class
-  plot_data = count(ddsb_filtered_hk_vehicles(), Vehicle_Class, name = "count", na.rm = TRUE) %>%
-    left_join(VEHICLE_CLASS_TRANSLATE, by = c("Vehicle_Class" = "Vehicle_Class")) %>%
+  plot_data = count(ddsb_filtered_hk_vehicles(), vehicle_class, name = "count", na.rm = TRUE) %>%
+    left_join(VEHICLE_CLASS_TRANSLATE, by = c("vehicle_class" = "vehicle_class")) %>%
     # Merge both en and zh values, then reorder vehicle class in descending order
-    mutate(Vehicle_Class_order = reorder(paste0(Vehicle_Class_chi, "\n", Vehicle_Class), count))
+    mutate(vehicle_class_order = reorder(paste0(vehicle_class_chi, "\n", vehicle_class), count))
 
-  plot_by_vehicle_class = ggplot(plot_data, aes(x = Vehicle_Class_order, y = count)) +
+  plot_by_vehicle_class = ggplot(plot_data, aes(x = vehicle_class_order, y = count)) +
     geom_bar(stat = "identity", fill = CATEGORY_COLOR$vehicles) +
     coord_flip() +
     theme_minimal() +
@@ -195,16 +195,16 @@ output$ddsb_all_vehicle_class_plot = renderPlotly({
 output$ddsb_all_road_hierarchy_plot = renderPlotly({
 
   # count by pedestrian Action
-  plot_data = ddsb_filtered_hk_accidents() %>%
-    filter(!is.na(Road_Hierarchy)) %>%
-    count(Road_Hierarchy, name = "count") %>%
-    left_join(ROAD_HIERARCHY_TRANSLATE, by = c("Road_Hierarchy" = "Road_Hierarchy")) %>%
+  plot_data = ddsb_filtered_hk_collisions() %>%
+    filter(!is.na(road_hierarchy)) %>%
+    count(road_hierarchy, name = "count") %>%
+    left_join(ROAD_HIERARCHY_TRANSLATE, by = c("road_hierarchy" = "road_hierarchy")) %>%
     # Merge both en and zh values, then reorder vehicle class in descending order
-    mutate(Road_Hierarchy_order = reorder(paste0(Road_Hierarchy_chi, "\n", Road_Hierarchy), count))
+    mutate(road_hierarchy_order = reorder(paste0(road_hierarchy_chi, "\n", road_hierarchy), count))
 
 
-  plot_by_road_hierarchy = ggplot(plot_data, aes(x = Road_Hierarchy_order, y = count)) +
-    geom_bar(stat = "identity", fill = CATEGORY_COLOR$accidents) +
+  plot_by_road_hierarchy = ggplot(plot_data, aes(x = road_hierarchy_order, y = count)) +
+    geom_bar(stat = "identity", fill = CATEGORY_COLOR$collisions) +
     coord_flip() +
     theme_minimal() +
     theme(