The coronavirus pandemic severely affected the lives of people and as a result lock down was imposed to contain the covid. These visualizations are presented so that people can make guided decisions based on the insights.
This map shows the total count of cases for confirmed, recovered, death and active cases. This can give a glimpse of some countries that have managed to eradicate the spread of virus.
This graph shows how rapidly cases are growing, reducing or are stable.
This graph is plotted to understand how the cases have grown overall. From this, we can get to know how effective are the measures undertaken to control the virus.
The negative plot of confirmed cases indicates the time where the control measures have been successful (recover rate).
The points close to x-axis indicates confirmed and recovery cases are similar. The depiction of plots should be stable rather than fluctuations. The plots being flat or moving towards better rate would mean countries doing well in managing the spread of virus.
GitHub. 2021. COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. [online] Available at: https://github.com/CSSEGISandData/COVID-19 [Accessed 9 June 2021].
Stack Overflow. 2021. [online] Available at: https://stackoverflow.com/ [Accessed 9 June 2021].
Plotly. 2021. [online] Available at: https://plotly.com/ [Accessed 9 June 2021].
Shiny R Studio. 2021. [online] Available at: https://shiny.rstudio.com/gallery/ [Accessed 9 June 2021].
GitHub. 2021. [online] Available at: https://github.com/RamiKrispin/coronavirus [Accessed 9 June 2021].
---
title: "Coronavirus Dashboard"
output:
flexdashboard::flex_dashboard:
orientation: rows
social: menu
source_code: embed
vertical_layout: fill
---
```{r setup, include=FALSE}
defaultWarn <- getOption("warn")
options(warn = -1)
#----------------------Packages---------------------------------
library(readr) # to read data
library(dplyr) # to prepare the data
library(ggplot2) # to plot static graphs
library(tidyr) # to manipulate and shape data
library(plotly) # to make graphs interactive
library(leaflet) # to plot dynamic map
library(flexdashboard) # to display dashboard
#----------------------Data-------------------------------------
# Reading the data
covid <- read.csv("https://raw.githubusercontent.com/RamiKrispin/coronavirus/master/csv/coronavirus.csv", stringsAsFactors = FALSE) %>%
mutate(date = as.Date(date, "%Y-%m-%d"),
country = factor(country, levels = unique(country)),
lat = as.numeric(lat),
long = as.numeric(long),
type = factor(type, levels = unique(type)))
# dropping province column and setting data types
covid <- select(covid, -matches("province"))
# New variable as month and year
covid$month <- format(covid$date, "%m")
covid$year <- format(covid$date, "%y")
#---- Wider format
covidW <- covid %>%
select(country, lat, long, type, cases) %>%
group_by(country, lat, long, type) %>%
summarise(cases = sum(cases, na.rm = TRUE)) %>%
pivot_wider(names_from = type,
values_from = cases) %>%
mutate(active = confirmed - death - recovered) %>%
arrange(-active)
#---- Longer format
covidL <-
covidW %>%
pivot_longer(cols = confirmed:active,
names_to = "type",
values_to = "cases")
# Global type cases
coW <- sum(as.numeric(covidW$confirmed), na.rm = TRUE)
reW <- sum(as.numeric(covidW$death), na.rm = TRUE)
deW <- sum(as.numeric(covidW$recovered), na.rm = TRUE)
acW <- sum(as.numeric(covidW$active), na.rm = TRUE)
# France cases.
covidFr <- covid %>% filter(country == "France")
# Daily new cases type for last 14 days in France. Also cumulative cases (sum of cases for each day)
frVis <- covidFr %>%
filter(date %in% tail(unique(as.Date(covidFr$date)), 14) ) %>%
select(date, type, cases) %>%
group_by(date, type) %>%
summarise(cases = sum(cases, na.rm = TRUE)) %>%
group_by(type) %>%
mutate(cumulative_cases = cumsum(cases))
# Daily new cases for Visualization
frDaily <- frVis %>%
select(date, type, cases) %>%
pivot_wider(names_from = type,
values_from = cases)
# Cumulative cases for Visualization
frCum <- frVis %>%
select(date, type, cumulative_cases) %>%
pivot_wider(names_from = type,
values_from = cumulative_cases) %>%
mutate(active = confirmed - death - recovered) %>%
mutate(active = cumsum(active))
# Growth rate for past 6 months
frGrowth <- covidFr %>% filter(year == "21") %>%
select(month, type, cases) %>%
group_by(month, type) %>% summarise(cases = sum(cases)) %>%
ungroup() %>%
group_by(type) %>% mutate(growth = ((cases - lag(cases))/lag(cases)) * 100) %>%
select(month, type, growth) %>%
pivot_wider(names_from = type,
values_from = growth)
# Recovery rate for past 6 months.
frRecovery <- covidFr %>%
filter(year == "21") %>%
select(month, type, cases) %>%
group_by(month, type) %>% summarise(cases = sum(cases)) %>%
ungroup() %>%
group_by(type) %>%
pivot_wider(names_from = type,
values_from = cases) %>%
mutate(recovery_rate = (recovered / confirmed) * 100) %>%
mutate(recovery_rate = if_else(is.na(recovery_rate), 0, recovery_rate))
# preparing for recovery rate visualization
frRecovery$rec_cri <- ifelse(frRecovery$recovery_rate <= 30 , frRecovery$recovery_rate, NA)
frRecovery$rec_avg <- ifelse(frRecovery$recovery_rate >= 31 && frRecovery$recovery_rate <= 60 , frRecovery$recovery_rate, NA)
frRecovery$rec_bet <- ifelse(frRecovery$recovery_rate >= 61 , frRecovery$recovery_rate, NA)
#-------------------------------Map-------------------------------------
# adding log cases for radius purpose
covidL <- covidL %>%
mutate(log_cases = 2 * log(cases))
# Split act as 3 data frames for 3 types
covidL.split <- covidL %>%
split(covidL$type)
# Map with different overlay
groupI <- names(covidL.split)
#-----------------------------Top 5 Countries---------------------------
# top 5 countries with most active cases
covCountry <- covidW %>%
ungroup() %>%
top_n(5, active)
# Extract those top 5 country names
countnames <- covCountry$country
# Contains top 5 countries with their daily cases
coworld <- covid %>%
select(date, country, type, cases) %>%
group_by(date, country, type) %>%
summarise(cases = sum(cases)) %>%
pivot_wider(names_from = type,
values_from = cases) %>%
mutate(active = confirmed - death - recovered) %>%
filter(country %in% countnames) %>%
arrange(date)
# Now considering top 5 countries with their daily cases for past 14 days
tailWorld <- coworld %>%
tail(70, date) %>%
select(date, country, confirmed) %>%
pivot_wider(names_from = country,
values_from = confirmed)
options(warn = defaultWarn)
```
Row
-----------------------------------------------------------------------
### confirmed {.value-box}
```{r}
valueBox(value = paste(format(coW, big.mark = ","), "", sep = " "),
caption = "Total Confirmed Cases",
icon = "fas fa-user-md",
color = "orange")
```
### active {.value-box}
```{r}
valueBox(value = paste(format(acW, big.mark = ","), "", sep = " "),
caption = "Total Active Cases", icon = "fas fa-ambulance",
color = "blue")
```
### recovered {.value-box}
```{r}
valueBox(value = paste(format(reW, big.mark = ","), "", sep = " "),
caption = "Total Recovered Cases", icon = "fas fa-heartbeat",
color = "green")
```
### death {.value-box}
```{r}
valueBox(value = paste(format(deW, big.mark = ","), "", sep = " "),
caption = "Total Death Cases",
icon = "fas fa-heart-broken",
color = "red")
```
Row {.tabset}
-----------------------------------------------------------------------
### Map
```{r}
# Map with different overlay circles
leaflet() %>% addTiles() %>% addProviderTiles(providers$CartoDB.Positron) %>%
addCircleMarkers(data = covidL.split[[1]], lat = ~lat, lng = ~long,
color = "blue",
stroke = FALSE,
fillOpacity = 0.5,
radius = ~log_cases,
popup = paste0("Country: ", covidL.split[[1]]$country,
"Type: ", covidL.split[[1]]$type,
"Cases: ", covidL.split[[1]]$cases),
group = groupI[1],
labelOptions = labelOptions(noHide = F,
direction = 'auto')) %>%
addCircleMarkers(data = covidL.split[[2]], lat = ~lat, lng = ~long,
color = "orange",
stroke = FALSE,
fillOpacity = 0.5,
radius = ~log_cases,
popup = paste0("Country: ", covidL.split[[2]]$country,
"Type: ", covidL.split[[2]]$type,
"Cases: ", covidL.split[[2]]$cases),
group = groupI[2],
labelOptions = labelOptions(noHide = F,
direction = 'auto')) %>%
addCircleMarkers(data = covidL.split[[3]], lat = ~lat, lng = ~long,
color = "red",
stroke = FALSE,
fillOpacity = 0.5,
radius = ~log_cases,
popup = paste0("Country: ", covidL.split[[3]]$country,
"Type: ", covidL.split[[3]]$type,
"Cases: ", covidL.split[[3]]$cases),
group = groupI[3],
labelOptions = labelOptions(noHide = F,
direction = 'auto')) %>%
addCircleMarkers(data = covidL.split[[4]], lat = ~lat, lng = ~long,
color = "green",
stroke = FALSE,
fillOpacity = 0.5,
radius = ~log_cases,
popup = paste0("Country: ", covidL.split[[4]]$country,
"Type: ", covidL.split[[4]]$type,
"Cases: ", covidL.split[[4]]$cases),
group = groupI[4],
labelOptions = labelOptions(noHide = F,
direction = 'auto')) %>%
addLayersControl(overlayGroups = names(covidL.split),
options = layersControlOptions(collapsed = FALSE))
```
### Top 5 countries with cases
```{r}
# Daily Confirmed Cases in Top 5 Countries for past 14 days (Log scale)
plot_ly(data = tailWorld,
x = ~ date,
y = ~ US,
name = 'US',
mode = 'lines+markers',
type = 'scatter',
stackgroup = 'one') %>%
add_trace(y = ~ France,
name = "France",
mode = 'lines+markers') %>%
add_trace(y = ~ `United Kingdom`,
name = "United Kingdom",
mode = 'lines+markers') %>%
add_trace(y = ~ Spain,
name = "Spain",
mode = 'lines+markers') %>%
add_trace(y = ~ `Netherlands`,
name = "Netherlands",
mode = 'lines+markers') %>%
layout(title = "Daily Confirmed Cases in Top 5 Countries for past 14 days (Log scale)",
yaxis = list(title = "Daily Confirmed Cases",
type = "log"),
xaxis = list(title = "Date",
type = "date"),
legend = list(x = 0.1, y = 0.9),
hovermode = "compare")
```
### Daily New Cases
```{r}
plot_ly(frDaily, x = ~ date, y = ~ confirmed, name = 'Confirmed', color = 'rgba(243, 156, 18, 1)',
type = 'scatter', mode = 'lines') %>%
add_trace(y = ~ recovered, name = "Recovered", color = 'rgba(46, 204, 113, 1)') %>%
add_trace(y = ~ death, name = "Death", color = 'rgba(240, 52, 52, 1)') %>%
layout(title = "Daily New Cases in France for past 14 days", yaxis = list(title = "Daily New Cases"),
xaxis = list(title = "Date", type = "date"), autosize = FALSE, legend = list(x = 0.1, y = 0.9), hovermode = "compare")
```
### Cumulative Cases
```{r}
plot_ly(frCum, x = ~ date, y = ~ confirmed, name = 'Confirmed', color = 'rgba(243, 156, 18, 1)',
type = 'scatter', mode = 'lines') %>%
add_trace(y = ~ recovered, name = "Recovered", color = 'rgba(46, 204, 113, 1)') %>%
add_trace(y = ~ death, name = "Death", color = 'rgba(240, 52, 52, 1)') %>%
layout(title = "Cumulative Cases in France for past 14 days", yaxis = list(title = "Cumulative Cases"),
xaxis = list(title = "Date", type = "date"), autosize = FALSE, legend = list(x = 0.1, y = 0.9), hovermode = "compare")
```
### Growth Rate
```{r}
plot_ly(frGrowth, x = ~month, y = ~ confirmed, name = 'Confirmed', color = 'rgba(46, 204, 113, 1)',
type = 'scatter', mode = 'lines+markers') %>%
add_trace(y = ~ recovered, name = "Recovered", color = 'rgba(240, 52, 52, 1)', type = 'scatter', mode = 'lines+markers') %>%
add_trace(y = ~ death, name = "Death", color = 'rgba(243, 156, 18, 1)', type = 'scatter', mode = 'lines+markers') %>%
layout(title = "Growth rate in France for past 5 months", yaxis = list(title = "Growth rate"),
xaxis = list(
ticktext = list("Feb", "March", "April", "May", "June"),
tickvals = list(2, 3, 4, 5, 6),
tickmode = "array",
title = "Months"
), autosize = FALSE, legend = list(x = 0.1, y = 0.9), hovermode = "compare")
```
### Recovery
```{r}
plot_ly(frRecovery, x = ~month, y = ~ round(rec_bet, 2), name = 'Better recovery rate', marker = list(color = 'green', size = 10),
type = 'scatter', mode = 'markers') %>%
add_trace(y = ~ round(rec_cri, 2), name = "Critical recovery rate", marker = list(color = 'red', size = 10),
type = 'scatter', mode = 'markers') %>%
add_trace(y = ~ round(rec_avg, 2), name = "Average recovery rate", marker = list(color = 'blue', size = 10),
type = 'scatter', mode = 'markers') %>%
layout(title = "Recovery rate in France for past 6 months", yaxis = list(title = "Recovery rate"),
xaxis = list(
ticktext = list("Jan", "Feb", "March", "April", "May", "June"),
tickvals = list(1, 2, 3, 4, 5, 6),
tickmode = "array",
title = "Months"
), autosize = FALSE)
```
### Summary
The coronavirus pandemic severely affected the lives of people and as a result lock down was imposed to contain the covid. These visualizations are presented so that people can make guided decisions based on the insights.
1. Map
This map shows the total count of cases for confirmed, recovered, death and active cases. This can give a glimpse of some countries that have managed to eradicate the spread of virus.
2. Daily New Cases
This graph shows how rapidly cases are growing, reducing or are stable.
3. Cumulative Cases
This graph is plotted to understand how the cases have grown overall. From this, we can get to know how effective are the measures undertaken to control the virus.
4. Growth Rate
The negative plot of confirmed cases indicates the time where the control measures have been successful (recover rate).
5. Recovery Rate
The points close to x-axis indicates confirmed and recovery cases are similar. The depiction of plots should be stable rather than fluctuations. The plots being flat or moving towards better rate would mean countries doing well in managing the spread of virus.
### References
* GitHub. 2021. COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. [online] Available at: [Accessed 9 June 2021].
* Stack Overflow. 2021. [online] Available at: [Accessed 9 June 2021].