server.R

library(shiny)
library(plotly)
library(shinyjs)
library(tidyr)
library(dplyr)
library(readr)
library(sm)



#reading all the necessary files to create the correct data frames
#the files are:
#child-mortality-by-sex.csv
#child-mortality.csv
#global-child-death-by-cause.csv
#global-child-mortality-timeseries.csv
#Children-woman-death-vs-survival.csv
raw.data <- read.csv("./data/child-mortality-by-sex.csv")
map.data <- read.csv("./data/child-mortality.csv", stringsAsFactors = FALSE)
causes.data <- read.csv("./data/global-child-deaths-by-cause.csv", stringsAsFactors = FALSE)
intro.data <- read.csv("./data/global-child-mortality-timeseries.csv", stringsAsFactors = FALSE)
data1 <- read.csv("./data/Children-woman-death-vs-survival.csv", stringsAsFactors = FALSE)

#This line removes all the rows which contains null values inthe male and female column
raw.data <- na.omit(raw.data)


#Renaming all the column names so as to make the column names much shorter and readable
#Raw data rename
colnames(raw.data)[1] <- "Country"
colnames(raw.data)[4] <- "Female-Ratio"
colnames(raw.data)[5] <- "Male-Ratio"
colnames(raw.data)[6] <- "Total-Population"
#Map data rename
colnames(map.data)[4] <- "MortalityRate"
#causes data rename
colnames(causes.data) <- c("Entity", "Code", "Year", "Deaths", "Population")


my.server <- function(input, output) {
  #A variable which keeps the track of whether the button was clicked or not
  counter <- reactiveValues(countervalue = 0)
  
  #This function, brings into view, another widget when the button is clicked and then hides it so that it can't be clicked anymore
  observeEvent(input$button, {
    shinyjs::show("ChildCountry2")
    shinyjs::hide("button")
    counter$countervalue <- counter$countervalue + 1 
  })
  
  #This creates a time series map that shows the status of the globalchild mortality rate
  output$worldmap <- renderPlotly({
    #removes all the years that weren't equal to the user input
    map.data <- map.data[map.data$Year == input$years, ]
    l <- list(color = toRGB("grey"), width = 0.5)
    
    # specify map projection/options
    g <- list(
      showframe = TRUE,
      showcoastlines = TRUE,
      projection = list(type = 'mollweide')
    )
    p <- plot_geo(map.data) %>%
      add_trace(
        z = ~map.data$MortalityRate, color = ~map.data$MortalityRate, colors = 'Reds',
        text = ~map.data$Entity, locations = ~map.data$Code, marker = list(line = l)
      ) %>%
      colorbar(title = 'Child-Mortality (per 1,000 live births)', ticksuffix = '%') %>%
      layout(
        title = 'Child Mortality',
        geo = g
      )
    return(p)
  })
  
  #Takes the user input of the country name and gender for eithe one or two countries and creates a scatter plot to show the child mortality rate
  #based on gender
  output$plot <- renderPlotly({
    #Getting the first country name
    result.one <- raw.data %>% filter(raw.data$Country == input$ChildCountry)
    #Getting the type of gender that needs to be displayed: Male/Female/Both
    gender.selection <- input$gender
    #Checks if the button was pressed and if it was takes in the second country name
    if(counter$countervalue > 0) {
      result.two <- raw.data %>% filter(raw.data$Country == input$ChildCountry2)
      if(gender.selection == "Male") {
        chart <- plot_ly(data = result.one, y = result.one$`Male-Ratio`, x = result.one$Year, type = 'scatter', mode = 'lines', name = input$ChildCountry)  %>% 
          add_trace( y = result.two$`Male-Ratio`, x = result.two$Year, name = input$ChildCountry2 ) 
      } else if(gender.selection == "Female") {
        chart <- plot_ly(data = result.one, y = result.one$`Female-Ratio`, x = result.one$Year, type = 'scatter', mode = 'lines', name = input$ChildCountry)  %>% 
          add_trace( y = result.two$`Female-Ratio`, x = result.two$Year, name = input$ChildCountry2 )
      } else {
        chart <- plot_ly(data = result.one, y = result.one$`Female-Ratio`, x = result.one$Year, type = 'scatter', mode = 'lines', name = paste( input$ChildCountry, "(Female)"))  %>% 
          add_trace( y = result.two$`Female-Ratio`, x = result.two$Year, name = paste( input$ChildCountry2, "(Female)") ) %>%
          add_trace( y = result.two$`Male-Ratio`, x = result.two$Year, name = paste( input$ChildCountry2, "(Male)")  )  %>%
          add_trace( y = result.one$`Male-Ratio`, x = result.one$Year, name = paste( input$ChildCountry, "(Male)") ) 
      }
    } else {
      if(gender.selection == "Male") {
        chart <- plot_ly(data = result.one, y = result.one$`Male-Ratio`, x = result.one$Year, type = 'scatter', mode = 'lines', name ="Country One")  
      } else if(gender.selection == "Female") {
        chart <- plot_ly(data = result.one, y = result.one$`Female-Ratio`, x = result.one$Year, type = 'scatter', mode = 'lines', name ="Country One")
      } else {
        chart <- plot_ly(data = result.one, y = result.one$`Female-Ratio`, x = result.one$Year, type = 'scatter', mode = 'lines', name ="Country One (Female)")  %>%
          add_trace( y = result.one$`Male-Ratio`, x = result.one$Year, name = "Country One (Male)" ) 
      }
    }
    return(chart)
  })
  
  #returns the brief overview of the data collected in the form of a table for country 1
  output$statitics1 <- renderTable ({
    result.one <- raw.data %>% filter(raw.data$Country == input$ChildCountry)
    statitics <- result.one
    return(statitics)
  })   
  
  #returns the brief overview of the data collected in the form of a table for country 2
  output$statitics2 <- renderTable ({
    statitics <- NULL
    if(counter$countervalue > 0) {
      result.two <- raw.data %>% filter(raw.data$Country == input$ChildCountry2)
      statitics <- result.two
    }
    return(statitics)
  })   
  
  #returns the name of certain diseases that lead to death among children
  output$diseases <- renderTable({
    df <- causes.data %>% select(Entity, Year, Deaths) %>% filter(Deaths >= 0)
    df <- spread(df, Year, Deaths) #converting it from long to wide
    return(df[1:14, "Entity"])
  })
  
  #header line
  output$table.info <- renderText({
    return("SUMMARY STATISTICS OF CHILD MORTALITY RATE BY GENDER")
  })
  
  #This returns the area graph showing informtion about what diseases lead to death among 
  #children and compares its effect from 1990 to 2015
  output$areaGraph <- renderPlotly({
    
    df <- causes.data %>% select(Entity, Year, Deaths) %>% filter(Deaths >= 0)
    df <- spread(df, Year, Deaths) #converting it from long to wide
    
    p <- plot_ly(x = df$Entity, y = df$'2015', type = 'scattergl', mode = 'lines', name = '2015', fill = 'tozeroy') %>%
      add_trace(x = df$Entity, y = df$'1990', name = '1990', fill = 'tozeroy') %>%
      layout(xaxis = list(title = "Leading Causes", showticklabels = FALSE),
             yaxis = list(title = 'Number of Deaths'), hovermode = 'compare')
    
    return(p)
  })
  
  
  #This is  a graph that we show on our introductory page which gives the user a brief understanding
  #about the child mortality rate in each country (based on the user's input)
  output$introPlot <- renderPlotly({
    #x axis will contain years based on user input
    #y axis will contain percent of population
    data <- intro.data %>% filter(Year >= input$introYears[1] & Year <= input$introYears[2])
    plot_ly(data, x = ~Year, y = ~Share.surviving.first.5.years.of.life....+ Share.dying.in.first.5.years.... , name = 'Share surviving ',
            type = 'scatter', mode = 'none', fill = 'tozeroy', fillcolor = '#F5FF8D',
            text = ~paste0(Year, '<br>Share surviving: ', Share.surviving.first.5.years.of.life....,
                           '<br>Share dying: ', Share.dying.in.first.5.years....)) %>%
      add_trace(y = ~Share.dying.in.first.5.years...., name = 'Share dying', fillcolor = '#50CB86') %>%
      
      layout(title = 'Global Child Mortality',
             xaxis = list(title = "",
                          showgrid = FALSE),
             yaxis = list(title = "Percent",
                          showgrid = FALSE,
                          ticksuffix = '%'))
    
  })

  #this returns an area graph compares the percentage of children that survived their 5ht birthday
  #and those who didnt based on a country
  output$rolePlot <- renderPlotly({
    #x axis will contain years based on user input
    #y axis will contain percent of population
    data2 <- data1 %>% filter(Year >= input$roleYears[1] & Year <= input$roleYears[2])  %>% filter(Entity == input$country)
    plot_ly(data2, x = ~Year, y = ~Children.that.survived.past.their.5th.birthday.per.woman..children.
            + Children.that.died.before.5.years.of.age.per.woman..Children.that.died.before.5.years.of.age.per.woman. ,
            name = 'died ', type = 'scatter', mode = 'none', fill = 'tozeroy', fillcolor = 'red',
            text = ~paste0(Year, ' ', Children.that.survived.past.their.5th.birthday.per.woman..children.
                           + Children.that.died.before.5.years.of.age.per.woman..Children.that.died.before.5.years.of.age.per.woman.,
                           '<br>Children that died before five years of age per woman: ', Children.that.died.before.5.years.of.age.per.woman..Children.that.died.before.5.years.of.age.per.woman.,
                           '<br>Children that survived after five years of age per woman: ',
                           Children.that.survived.past.their.5th.birthday.per.woman..children.)) %>%
      add_trace(y = ~Children.that.survived.past.their.5th.birthday.per.woman..children., name = 'survived', fillcolor = '#50CB86') %>%
      
      layout(title = 'How many children did a woman give birth to that died before their 5th birthday?',
             xaxis = list(title = "",
                          showgrid = FALSE),
             yaxis = list(title = "Children per woman",
                          showgrid = FALSE))
    
  })
  
  #bried summary about the graph regarding infant mortality
  output$fertility.intro <- renderText({
    return("This visualization represents the total number of children a woman gave birth to that were victims of child 
            mortality (died before the age of 5). One can view the contrast regarding different countries, 
            i.e. the number of children lost by a woman in different countries over time")
  })
  
  #bried summary about the graph regarding infant mortality
  output$fertility.summary <- renderText({
    return("The slider allows the user to select the time period and the drop-down menu lets the user select the country. 
            The line density plot then displays the comparison between the number of children that survived above the age
            of 5 (per woman) vs. the number children that did not survive above the age of 5 (per woman). 
            The Y-axis represents the frequency of children per woman and the X-axis represents the year span")
  })
  
  #introductory paragraph
  output$intro <- renderText({
    return("Child Mortality has been a social issue that has been ever-present in the society. Simply defined, it is the 
           rate of deaths of children below 5 years of age. Child mortality is dependent on a number of factors varying 
           from parents' biological wellness to proper health care facilities being provided. As per data provided by 
           UNICEF, child mortality rates around the world have been on a decline in the recent years. In 19th century 
           central Europe,  specifically Germany, the child mortality rate was 1 in 2 births whereas in today's 
           Germany the child mortality rate has significantly come down to 3.7 in 1000 live births. 
           Even on a global scale, the child mortality rate has reduced to 3.7 million from a massive 7.6 million in the 
           last 2 decades. A positive development on such a serious social issue has certainly not been highlighted 
           publically by the media. It is largely due to the fact that the change on global level is a sedate process, 
           but is it definitely something that's informative and can prove insightful in taking future steps. 
           This shiny app has been created to primarily achieve the goal of extracting and portraying valuable 
           information for NGO's and health organizations around the world working towards improving child mortality 
           rates in different countries. Child Mortality rate data ranging from 1751 to 2015 has been intuitively depicted 
           through this app which gives an idea of the current state of healthcare facilities andd could consequesntly help
           take better decisions regarding policies in the future.")
  })
  
  #bried summary about the graph regarding causes of child mortality
  output$causes <- renderText({
    return("This visualization displays the global freqeuncy of child deaths and their causes in the form of an area-filled
            graph. The frequency of deaths are on the Y-axis whereas their respective causes are on the X-axis. This visualization
             uses multiple traces which means that is compares the cases with respect to two cases, i.e. 1990 and 2015
             For example: The highest deaths recorded in 1990 due to lower respiratory infection were 
             2.12 million deaths. Correspondingly, the number of deaths recorded due to the same cause in 
             2015 were 703.9K deaths. We can clearly infer that the child mortality rate was higher in 
             1990 than 2015. Since the visulaization tells us that more than half of all(51.8% to be precise) 
             children died due to infectious diseases, we can conclude that over the years there has been an improvement
              in the availability of vaccinations. ")
  })
  
  #bried summary about the graph regarding global child mortality
  output$map.info <- renderText({
    return("In the time series plot we try to showcase child mortality over the long run. Today child mortality in industrialized 
           countries is below 0.5%. The time series plot shows that these low mortality rates are a very recent development, 
           and in pre-modern countries child mortality rates were between 30% and 50%. In the late 19th century, every second 
           child in Germany died before its fifth birthday. In developing countries the health of children is quickly improving 
           – but child mortality is still much higher than in developed countries")
  })
  
  #bried summary about the graph regarding child mortality based on gender
  output$gender <- renderText({
    return("This visualization represents the child mortality rate in the form of a line plot. One can select a country
             and view the child mortality rates of boys or girls or even compare the child mortality rates between
            genders and different countries. On the Y-axis is the child mortality rate and on the X-axis is the year 
            ranging from 1990 to 2010. Looking at a fair number of instances, we can infer that in most countries, 
            over the recent years, the child mortality rates within boys has consistenty been greater than that of girls.
           ")
  })
  
  #Inference about the visualisation behind the graph talking about causes of child mortality
  output$causes.inference <- renderText({
    return("INFERENCE: From this data visualization, it is easy to point out which diseases in the current day and age are the
           major reasons behind the death of children and thus appropritae actions can be taken to fight them, including but 
           not limited to : 1) spreading awareness about how to prevent the spread of the diseases
                            2) openning up vacinnation stalls in places in the most effected places")
  })
  
  #Inference: time series world map
  output$world.inderence <- renderText({
    return("INFERENCE: From this data visualization, it is easy to point out which countries still have a major problem
           regarding child mortality and need immediate assistance. As a result, appropriate actions can be taken quickly
           so as to prevent further harm from happening. This map also manages to tell us that as countries begin to progress,
           the rate of child mortality decreaes accordingly. This tells us how important it is to improve the living standards
           in order to tackle this dreadful problem")
  })
  
}

shinyServer(my.server)