{r setup, include=FALSE} knitr::opts_chunk$set(cache = TRUE, echo = FALSE, message = FALSE, warning = FALSE)

Introduction

This project was created as part of the Developing Data Products course of the Coursera Data Science Specialisation. The goal of the project is to create a web page using R Markdown that features a map created with Leaflet, and to host the resulting web page on either GitHub Pages, RPubs, or NeoCities.

Note - The R Markdown source code of this presentation can be found in this GitHub repository.

Map of passenger traffic

The following map represents passenger traffic on the Paris RATP railway network in 2015. Clicking on a station shows the total number of passengers that entered the station during that year.

{r prerequisites} rm(list=ls()) library(leaflet) library(jsonlite) library(dplyr) library(stringr)

{r load_data} # load traffic data station_traffic <- read.csv2( "data/trafic-annuel-entrant-par-station-du-reseau-ferre-2015.csv", header = TRUE, encoding = "UTF-8", stringsAsFactors = FALSE) # keep only columns of interest (network, station, traffic) station_traffic <- station_traffic %>% select(Réseau, Station, Trafic) # load location data station_location <- read.csv2( "data/emplacement-des-gares-idf-data-generalisee.csv", header = TRUE, encoding = "UTF-8", stringsAsFactors = FALSE) # load geographical network layout data geocsv <- read.csv2("data/traces-du-reseau-ferre-idf.csv", stringsAsFactors = FALSE)

{r prepare_traffic_data} # keep only columns of interest (coordinates, long name, line types) station_location <- station_location %>% select(Geo.Point, NOM_LONG, RER, METRO) # split Geo.Point string into latitude and longitude station_lat_lng <- data.frame( str_split_fixed(station_location$Geo.Point, ", ", 2), stringsAsFactors = FALSE) colnames(station_lat_lng) <- c("lat", "lng") # convert coordinates to numeric station_lat_lng <- sapply(station_lat_lng, as.numeric) # replace Geo.Point with separate coordinates station_location <- cbind(station_lat_lng, station_location) station_location <- station_location %>% select(-Geo.Point) # remove two RER stations on the SNCF network that have the same name as # metro (RATP) stations; only the latter are needed ### found using: ## station_location[duplicated(station_location$NOM_LONG), "NOM_LONG"] station_location <- station_location[ !(station_location$NOM_LONG %in% c("MALESHERBES", "SAINT-FARGEAU") & station_location$RER == 1), ] # create vector mapping names of stations from the traffic data set to # those of the location data set, to be used to join the two data sets stationKeyName <- station_traffic$Station # drop "-RER" suffix from stations that are both metro and RER stations stationKeyName <- gsub("-RER$", "", stationKeyName) # manually define mapping for 54 stations that are named differently in the # two data sets stationKeyName <- sub("^GARE D'AUSTERLITZ", "AUSTERLITZ", stationKeyName) stationKeyName <- sub("^VAUGIRARD", "VAUGIRARD (ADOLPHE CHERIOU)", stationKeyName) stationKeyName <- sub("^ETIENNE MARCEL", "ETIENNE-MARCEL", stationKeyName) stationKeyName <- sub("^PLACE MONGE", "PLACE MONGE (JARDIN DES PLANTES)", stationKeyName) stationKeyName <- sub("^BEL AIR", "BEL-AIR", stationKeyName) stationKeyName <- sub("^LES AGNETTES", "LES AGNETTES-ASNIERES-GENNEVILLIERS", stationKeyName) stationKeyName <- sub("^LOUVRE", "LOUVRE-RIVOLI", stationKeyName) stationKeyName <- sub("^LA COURNEUVE-8 MAI 1945", "LA COURNEUVE 8 MAI 1945", stationKeyName) stationKeyName <- sub("^CRETEIL-PREFECTURE", "CRETEIL-PREFECTURE (HOTEL DE VILLE)", stationKeyName) stationKeyName <- sub("^PEREIRE", "PEREIRE-LEVALLOIS", stationKeyName) stationKeyName <- sub("^MARNE-LA-VALLEE-CHESSY", "CHESSY - MARNE-LA-VALLEE", stationKeyName) stationKeyName <- sub("^VAL D'EUROPE", "VAL-D'EUROPE", stationKeyName) stationKeyName <- sub("^VOLTAIRE", "VOLTAIRE (LEON BLUM)", stationKeyName) stationKeyName <- sub("^LES SABLONS", "LES SABLONS (JARDIN D'ACCLIMATATION)", stationKeyName) stationKeyName <- sub("^BOULOGNE-PONT DE SAINT-CLOUD", "BOULOGNE-PONT DE SAINT CLOUD", stationKeyName) stationKeyName <- sub("^CORENTIN CELTON", "CORENTIN-CELTON", stationKeyName) stationKeyName <- sub("^JAVEL-ANDRE CITROEN", "JAVEL", stationKeyName) stationKeyName <- sub("^BOBIGNY-PANTIN-RAYMOND QUENEAU", "BOBIGNY-PANTIN (RAYMOND QUENEAU)", stationKeyName) stationKeyName <- sub("^LA TOUR-MAUBOURG", "LA TOUR MAUBOURG", stationKeyName) stationKeyName <- sub("^PONT NEUF", "PONT-NEUF (LA MONNAIE)", stationKeyName) stationKeyName <- sub("^PARC DE SCEAUX", "PARC-DE-SCEAUX", stationKeyName) stationKeyName <- sub("^LE PARC DE SAINT-MAUR", "LE PARC-DE-SAINT-MAUR", stationKeyName) stationKeyName <- sub("^FRANKLIN D. ROOSEVELT", "FRANKLIN-D.ROOSEVELT", stationKeyName) stationKeyName <- sub("^GRANDS BOULEVARDS", "GRANDS BOULEVARDS (RUE MONTMARTRE)", stationKeyName) stationKeyName <- sub("^PALAIS-ROYAL", "PALAIS ROYAL-MUSEE DU LOUVRE", stationKeyName) stationKeyName <- sub("^LA DEFENSE", "LA DEFENSE-GRANDE ARCHE", stationKeyName) stationKeyName <- sub("^MONTPARNASSE-BIENVENUE", "MONTPARNASSE", stationKeyName) stationKeyName <- sub("^CHAUSSEE D'ANTIN-LA FAYETTE", "CHAUSSEE D'ANTIN (LA FAYETTE)", stationKeyName) stationKeyName <- sub("^PLACE CLICHY", "PLACE DE CLICHY", stationKeyName) stationKeyName <- sub("^AUBERVILLIERS-PANTIN-QUATRE CHEMINS", "AUBERVILLIERS-PANTIN (QUATRE CHEMINS)", stationKeyName) stationKeyName <- sub("^FUNICULAIRE", "FUNICULAIRE MONTMARTRE STATION BASSE", stationKeyName) stationKeyName <- sub("^LA CROIX DE BERNY", "LA CROIX-DE-BERNY-FRESNES", stationKeyName) stationKeyName <- sub("^SAINT-GERMAIN DES PRES", "SAINT-GERMAIN-DES-PRES", stationKeyName) stationKeyName <- sub("^GABRIEL PERI", "GABRIEL PERI-ASNIERES-GENNEVILLIERS", stationKeyName) stationKeyName <- sub("^GALLIENI", "GALLIENI (PARC DE BAGNOLET)", stationKeyName) stationKeyName <- sub("^CHARLES DE GAULLE-ETOILE", "CHARLES DE GAULLE ETOILE", stationKeyName) stationKeyName <- sub("^PRE-SAINT-GERVAIS", "PRE SAINT-GERVAIS", stationKeyName) stationKeyName <- sub("^SAINT-MICHEL-NOTRE-DAME", "SAINT-MICHEL", stationKeyName) stationKeyName <- sub("^CLUNY LA SORBONNE", "CLUNY-LA-SORBONNE", stationKeyName) stationKeyName <- sub("^RUE DE LA POMPE", "RUE DE LA POMPE (AVENUE GEORGES MANDEL)", stationKeyName) stationKeyName <- sub("^BIBLIOTHEQUE", "BIBLIOTHEQUE FRANCOIS MITTERRAND", stationKeyName) stationKeyName <- sub("^PONT DE NEUILLY", "PONT DE NEUILLY (AVENUE DE MADRID)", stationKeyName) stationKeyName <- sub("^GONCOURT", "GONCOURT (HOPITAL SAINT-LOUIS)", stationKeyName) stationKeyName <- sub("^QUATRE-SEPTEMBRE", "QUATRE SEPTEMBRE", stationKeyName) stationKeyName <- sub("^CARDINAL LEMOINE", "CARDINAL-LEMOINE", stationKeyName) stationKeyName <- sub("^PONT MARIE", "PONT-MARIE (CITE DES ARTS)", stationKeyName) stationKeyName <- sub("^SAINT-SEBASTIEN-FROISSART", "SAINT-SEBASTIEN FROISSART", stationKeyName) stationKeyName <- sub("^MALAKOFF-RUE ETIENNE DOLET", "MALAKOFF-RUE TIENNE DOLET", stationKeyName) stationKeyName <- sub("^CRETEIL-L'ECHAT", "CRETEIL-L'ECHAT (HOPITAL MONDOR)", stationKeyName) stationKeyName <- sub("^LES COURTILLES", "LES COURTILLES-ASNIERES-GENNEVILLIERS", stationKeyName) stationKeyName <- sub("^CROIX DE CHAVAUX", "CROIX DE CHAVAUX (JACQUES DUCLOS)", stationKeyName) stationKeyName <- sub("^VAL DE FONTENAY", "VAL-DE-FONTENAY", stationKeyName) stationKeyName <- sub("^BIR-HAKEIM", "BIR-HAKEIM (GRENELLE)", stationKeyName) stationKeyName <- sub("^SAINT-PAUL", "SAINT-PAUL (LE MARAIS)", stationKeyName) # add the mapped names to the traffic data set station_traffic$keyName <- stationKeyName # join the two data sets (removes one empty row) station_all <- inner_join(station_traffic, station_location, by = c("keyName" = "NOM_LONG")) # add jitter to coordinates of overlapping stations (i.e. with same key name) dup_stations <- station_all$keyName[duplicated(station_all$keyName)] is_dup_station <- station_all$keyName %in% dup_stations set.seed(12345) station_all$lat[is_dup_station] <- jitter(station_all$lat[is_dup_station], factor = .1) station_all$lng[is_dup_station] <- jitter(station_all$lng[is_dup_station], factor = .1) # sort stations in order of decreasing traffic so that smaller markers are # displayed on top of larger ones station_all <- station_all[order(station_all$Trafic, decreasing = TRUE), ]

```{r prepare_network_layout_data} # GeoJSON helper functions # FUNCTION # genFeatureCollectionFromGeoShapeList(shapes, style) # # DESCRIPTION # Converts a list of GeoJSON shapes (geometries) to a single GeoJSON # FeatureCollection that can be passed to leaflet’s addGeoJson() function # # INPUT # - shapes: a list of GeoJSON shapes # - style: a style list to apply to the FeatureCollection # # OUTPUT # A GeoJSON FeatureCollection that can be passed to leaflet’s addGeoJSON() # function genFeatureCollectionFromGeoShapeList <- function(shapes, style) { features <- lapply(shapes, function(shape) { feature <- list() feature\(type <- "Feature" feature\)geometry <- fromJSON(shape) feature\(properties <- list() return(feature) }) featureCollection <- list() featureCollection\)type <- “FeatureCollection” featureCollection\(features <- features featureCollection\)style = style

return(featureCollection) } # FUNCTION # genFeatureCollectionFromGeoCsv(gcsv, nwk_com, colour, weight) # # DESCRIPTION # Extracts a list of GeoJSON shapes (geometries) for a given line of the # layout of the Île-de-France railway network data set, and returns a styled # GeoJSON FeatureCollection that can be passed to leaflet’s addGeoJson() # function # # INPUT # - gcsv: a data frame containing the layout of the Île-de-France railway # network data set # - nwk_com: a commercial line name (e.g. “M1” for metro line 1), # corresponding to the RES_COM column of the aforementioned data set # - colour: the colour of the line (e.g. “#FFCD00” for metro line 1) # - weight: the weight to be used to style the line # # OUTPUT # A GeoJSON FeatureCollection that can be passed to leaflet’s addGeoJSON() # function genFeatureCollectionFromGeoCsv <- function(gcsv, nwk_com, colour, weight) { return(genFeatureCollectionFromGeoShapeList( gcsv\(Geo.Shape[gcsv\)RES_COM == nwk_com], list( weight = weight, color = colour, opacity = 1, fillOpacity = 0 ) )) } ```

{r geojson} # generate FeatureCollection for each line (metro and RERs A and B) m1 <- genFeatureCollectionFromGeoCsv(geocsv, "M1", "#FFCD00", 3) m2 <- genFeatureCollectionFromGeoCsv(geocsv, "M2", "#003CA6", 3) m3 <- genFeatureCollectionFromGeoCsv(geocsv, "M3", "#837902", 3) m3b <- genFeatureCollectionFromGeoCsv(geocsv, "M3bis", "#6EC4E8", 3) m4 <- genFeatureCollectionFromGeoCsv(geocsv, "M4", "#CF009E", 3) m5 <- genFeatureCollectionFromGeoCsv(geocsv, "M5", "#FF7E2E", 3) m6 <- genFeatureCollectionFromGeoCsv(geocsv, "M6", "#6ECA97", 3) m7 <- genFeatureCollectionFromGeoCsv(geocsv, "M7", "#FA9ABA", 3) m7b <- genFeatureCollectionFromGeoCsv(geocsv, "M7bis", "#6ECA97", 3) m8 <- genFeatureCollectionFromGeoCsv(geocsv, "M8", "#E19BDF", 3) m9 <- genFeatureCollectionFromGeoCsv(geocsv, "M9", "#B6BD00", 3) m10 <- genFeatureCollectionFromGeoCsv(geocsv, "M10", "#C9910D", 3) m11 <- genFeatureCollectionFromGeoCsv(geocsv, "M11", "#704B1C", 3) m12 <- genFeatureCollectionFromGeoCsv(geocsv, "M12", "#007852", 3) m13 <- genFeatureCollectionFromGeoCsv(geocsv, "M13", "#6EC4E8", 3) m14 <- genFeatureCollectionFromGeoCsv(geocsv, "M14", "#62259D", 3) rerA <- genFeatureCollectionFromGeoCsv(geocsv, "RER A", "#E2231A", 5) rerB <- genFeatureCollectionFromGeoCsv(geocsv, "RER B", "#7BA3DC", 5)

```{r map} # define palette for markers (green for metro network, blue for RER network) pal <- colorFactor(c(“#00C4B3”, “#050D9E”), domain = c(“Métro”, “RER”)) # create and display map m <- station_all %>% leaflet() %>% addProviderTiles(“CartoDB.Positron”) %>% setView(lng = 2.352, lat = 48.857, zoom = 12) %>%

# add RER and metro lines addGeoJSON(rerA) %>% addGeoJSON(rerB) %>% addGeoJSON(m1) %>% addGeoJSON(m2) %>% addGeoJSON(m3) %>% addGeoJSON(m3b) %>% addGeoJSON(m4) %>% addGeoJSON(m5) %>% addGeoJSON(m6) %>% addGeoJSON(m7) %>% addGeoJSON(m7b) %>% addGeoJSON(m8) %>% addGeoJSON(m9) %>% addGeoJSON(m10) %>% addGeoJSON(m11) %>% addGeoJSON(m12) %>% addGeoJSON(m13) %>% addGeoJSON(m14) %>%

# add stations (size represents traffic) addCircleMarkers(~lng, ~lat, radius = ~sqrt(Trafic/100000), color = ~pal(Réseau), stroke = FALSE, fillOpacity = 1, popup = ~paste(“”, Station, “
”, format(Trafic, big.mark = “,”, trim = TRUE), sep = “”)) m ```

Data sources and licences

This project uses the following data: