Extracting example data

I have used Yelp data for this example, as it is ideal as it has geographic information, along with additional data to create a plot for each location. This was extracted pro grammatically in R using the Yelp API.

In order to use the Yelp API, you need to create an app through their development area. This allows you to create a client id and api key which you can use to extract data.

I create the following function, which relies on the httr and jsonlite packages to extract the data, along with the tidyverse to clean it. It requires 6 inputs:

• term - what are you searching for

• location - the area where you want to search

• limit - how many results do you want to find

• radius - radius of the location

• client_id - client id key from Yelp

• api - api key from Yelp

library(httr)
library(jsonlite)
library(tidyverse) 

yelp_search<-function(term,location,limit, radius,client_id,api){
  yelp <- "https://api.yelp.com"

  url <- modify_url(yelp, path = c("v3", "businesses", "search"),
                    query = list(term = term, location = location, 
                                 limit = limit,
                                 radius = radius))
  res <- GET(url, add_headers('Authorization' = paste("bearer", api)))
  
  resTEXT<-httr::content(res, as="text")
  JLres<-jsonlite::fromJSON(resTEXT, flatten=TRUE)
  
  BUS<-JLres$businesses
  cat_list<-list()
  for (i in 1:length(BUS$name)){
    categories1<-BUS$categories[[i]] %>%
      as.vector()
    c2<-as.vector(categories1$title)
    c3<-paste0(c2, collapse = ", ")
    cat_list[[i]]<-c3
    }
    
  c4<-unlist(cat_list)
  

  BUSINESS_DF<-select(BUS,name,rating,review_count,price,phone,
                      latitude=coordinates.latitude,
                      longitude=coordinates.longitude,
                      postcode=location.zip_code)%>%
    mutate(categories=c4)%>%
    as_tibble()
  
  BUSINESS_DF$price_num<-nchar(BUSINESS_DF$price)
  
  return(BUSINESS_DF)
  
  }

In this example we are extracting data for barber shops in Manchester, UK. I filter, only examining the data where there is price information.

term <- "barber"
location <- "Manchester, UK"
limit <- 20
radius <- 8000
#api and client_id are obtained from Yelp developer page

yelp_df <- yelp_search(term,location,limit,radius,client_id,api)%>%
  filter(!is.na(price_num))

Map with markers

the first thing that you can do with googleway is to map the data on google maps with markers.

You can include additional information that will appear when you click on the marker. To do this we add an extra column to the dataframe with the info that will appear when you click on the marker. In this case it is the name and rating.

yelp_df$info<-paste0("<b>Name: </b>",yelp_df$name," ",
                "<b>Rating: </b>",yelp_df$rating)

We then plot this data with googleway. However, before using googleway you need to register with google maps and obtain an api key, details on how to do this are provided here

In this example we refer to the google map api as key.

library(googleway)
library(tidyverse)

set_key(key = key) #key is the api

google_map() %>%
  add_markers(
    data = yelp_df,
    info_window = "info"
  )

You can also alter the colours of the markers, instructions on how to do this can be found here

Map with plots

To replace the markers with plots, there are a number of steps we have to take:

• Create the plots with ggplot
  
• Save plots to a file and create a link for the file
  
• Re-plot with googleway

library(tidyverse)
library(reshape2)
for(i in 1:length(yelp_df$name)) {
  name_place<-yelp_df$name[[i]]
  yp<-select(yelp_df,name,rating,review_count,price_num)%>%
    melt()
  p <- ggplot(subset(yp, name==name_place), 
              aes(variable, value,  fill = variable)) + 
    geom_bar(stat="identity", show.legend=FALSE)+
    theme(axis.title.x=element_blank(),
          axis.text.x=element_blank(),
          axis.text.y=element_blank(),
          axis.ticks.y=element_blank(),
          axis.ticks.x=element_blank(),
          axis.title.y=element_blank(),
          panel.background = element_rect(fill = "transparent"), # bg of the panel
          plot.background = element_rect(fill = "transparent", color = NA), # bg of the plot
          panel.grid.major = element_blank(), # get rid of major grid
          panel.grid.minor = element_blank(), # get rid of minor grid
          legend.background = element_rect(fill = "transparent"), # get rid of legend bg
          legend.box.background = element_rect(fill = "transparent") # get rid of legend panel bg
)
  file_name<-paste0("figure_",i,"_",name_place,".png")%>%
    tolower()%>%
    gsub("[[:space:]]", "", .)
  
  ggsave(file_name, p,width=0.6,height=0.3,
           bg = "transparent")
}

path_to_images <-"https://raw.githubusercontent.com/matthewsmith430/vis_image/master/"

for (i in 1:length(yelp_df$name)){
  name_place<-yelp_df$name[[i]]
  
  
  file_name<-paste0("figure_",i,"_",name_place,".png")%>%
    tolower()%>%
    gsub("[[:space:]]", "", .)
  
  yelp_df$image[i]<-paste0(path_to_images, file_name)
  
}

library(googleway)
library(tidyverse)

set_key(key = key) #key is the api

google_map() %>%
  add_markers(
    data = yelp_df,
    marker_icon = "image",
    info_window = "info"
  )