Visualize the Hubway Data

hubway_dataviz.R

##################################################################################################################
######
######    Code to visualize the Hubway Data Set 
######    for the Hubway Data Challenge
######    http://hubwaydatachallenge.org/
######
######    Jarrett Byrnes, http://jarrettbyrnes.info
######
######    Last Updated Nov 5, 2012
##################################################################################################################


##
# Load some libraries
##
library(plyr)
library(ggplot2)
library(reshape2)

## 
#Load the Data
##
trips <- read.csv("./trips_time.csv")
stations <- read.csv("./stations.csv")

latLong_lookup <-  function(id) {
  idx <- match(id, stations$id)
  return(data.frame(lat = stations$lat[idx], lng = stations$lng[idx]))
}

##
# extract times
## 
trips$start_date<- as.POSIXlt(trips$start_date)
trips$end_date<- as.POSIXlt(trips$end_date)
trips$start_Time <- format(trips$start_date, format="%H:00")
trips$end_Time <- format(trips$end_date, format="%H:00")


#############
#get the information about all of the trips ever
#############

trips_aggregated_all <- ddply(trips, .(start_station_id, end_station_id), 
                          summarize, n.Trips = length(duration), mean.duration = mean(duration, na.rm=T))

trips_aggregated_all<- cbind( trips_aggregated_all,
							latLong_lookup(trips_aggregated_all$start_station_id),
							latLong_lookup(trips_aggregated_all$end_station_id))

names(trips_aggregated_all)[7:8]<-paste(names(trips_aggregated_all)[7:8], ".end", sep="")


#plot it!
ggplot(trips_aggregated_all, aes(x=lng, y=lat, 
							xend=lng.end, yend=lat.end, 
							color=log10(n.Trips), size=mean.duration)) +
							geom_segment(alpha=0.03) +
							theme_bw() +
							xlab("Longitude") + ylab("Latitude") +
							scale_color_gradient("Log of Number of Trips", 
										low="blue", high="red") +
							scale_size_continuous("Trip Duration") +
							 guides(size = guide_legend(override.aes = list(alpha = 1))) +
							 geom_point(data=stations, 
							 aes(x=lng, y=lat, xend=lng, yend=lat), size=1,color="black")


ggsave("hubway_alltraffic.pdf")


########################################
#####Now aggregate by start time
########################################
trips_aggregated_start <- ddply(trips, .(start_station_id, end_station_id, start_Time),
						 summarize,
                         n.Trips = length(duration), mean.duration = mean(duration, na.rm=T))

trips_aggregated_start<- cbind( trips_aggregated_start,
							latLong_lookup(trips_aggregated_start$start_station_id),
							latLong_lookup(trips_aggregated_start$end_station_id))
names(trips_aggregated_start)[8:9]<-paste(names(trips_aggregated_start)[8:9], ".end", sep="")

ggplot(trips_aggregated_start, aes(x=lng, y=lat, 
							xend=lng.end, yend=lat.end, 
							color=log(n.Trips), size=mean.duration)) +
							geom_segment(mapping=aes(alpha=n.Trips)) +
							 facet_wrap(~ start_Time)+
							theme_bw() +
							xlab("Longitude") + ylab("Latitude") +
							scale_color_gradient("Log of Number of Trips", 
										low="blue", high="red") +
							scale_size_continuous("Trip Duration") +
							 guides(size = guide_legend(override.aes = list(alpha = 1))) +
							 geom_point(data=stations, 
							 aes(x=lng, y=lat, xend=lng, yend=lat), size=1,color="black") +
							 ggtitle("Trips Plotted by Start Time \n") +
							 scale_alpha_continuous("Number of Trips", range=c(0.01, 0.05))+
							 scale_x_continuous(breaks=c(-71.15, -71.1, -71.05)) +
							 scale_y_continuous(breaks=c(42.31, 42.35, 42.37))
						
							 
ggsave("trips_by_start_time.pdf")