Saleem Burhani - Karachi - R Assigment 2

Saleem Burhani - Karachi - assignment-2.R

@@ -0,0 +1,240 @@
+# R Assignment 2
+# 22 March 2017
+# Completed 27 march 2017
+# Saleem Burhani
+
+# load libraries
+library(swirl)
+#onetime usage
+#swirl::install_course("Getting and Cleaning Data")
+#library(datasets)
+library(dplyr)
+#library(sparklyr)
+#library(tidyr)
+library(lubridate)
+library(ggplot2)
+library(stringr)
+
+# load dataset
+path2csv = "C:\\Users\\biuser\\Desktop\\DIH\\22mar17-R&P\\hospitaldata.csv"
+mydfCSV <- read.csv(path2csv,stringsAsFactors=FALSE)
+
+
+
+"Task ........1"
+"Please remove the dots in the names, so it may become easier for you to work through it."
+" "
+# Name column is 
+mydfcolnames <- colnames(mydfCSV)
+mydfcolnames <- gsub(".."," ",mydfcolnames,fixed=TRUE)
+mydfcolnames <- gsub("."," ",mydfcolnames,fixed=TRUE)
+colnames(mydfCSV) <- mydfcolnames
+head(mydfCSV)
+
+# Remove all "Nursing Staff" records which are not required
+mydf <- filter(mydfCSV,mydfCSV$`Consulting Doctor` != "Nursing Staff")
+
+"Task 2"
+"Which day of the week is expected to have most visits?"
+" "
+
+# create column have week day
+mydf <- mutate(mydf,Week_Day = substr(mydf$Date,0,str_locate(mydf$Date,",")-1))
+# count records group by week_day 
+patient_by_week_day <- aggregate(rep(1, length(mydf$Week_Day)),by=list(mydf$Week_Day), sum)
+# get the day having max records
+max_patient_in_a_week_day <- max(patient_by_week_day$x, na.rm = TRUE)
+filter(patient_by_week_day, x == max_patient_in_a_week_day)
+
+"Task 3"
+"What is the average age of patients?"
+" "
+newdf <- mydf
+newdf$age <- as.numeric(mydf$Age, na.rm=TRUE)
+mean(newdf$age, na.rm=TRUE)
+
+
+"Task 4"
+"How many children were entertained? (Make a Bracket of Age from 1-12)"
+" "
+newdf <- mutate(newdf, adult_child = ifelse (age>=1 & age<=12,"Child","Adult"))
+count(newdf,adult_child)
+
+"Task 5"
+"Which gender type had what kind of procedure in abundance? i.e. Female visit mostly because of Gynae Problem"
+" "
+#filter(newdf, Sex == "M")
+temp <- count(newdf,Sex,Specialty,sort=TRUE)
+tempM <- filter(temp,Sex == "M")
+tempF <- filter(temp,Sex == "F")
+tempM[1,]
+tempF[1,]
+#max(tempM$n, na.rm=TRUE)
+#max(temp$n, na.rm = TRUE)
+
+"Task 6"
+"Which Doctor is earning highest?"
+" "
+newdf$`Amount Received ` <- as.numeric(mydf$`Amount Received `, na.rm=TRUE)
+doctors <- aggregate(newdf$`Amount Received `, by=list(Doctor=newdf$`Consulting Doctor`), FUN=sum)
+filter(doctors, x == max(doctors$x, na.rm=TRUE))
+
+"Task 7"
+"Which procedure type earns more money?"
+" "
+Procedures <- aggregate(newdf$`Amount Received `, by=list(Procedures=newdf$Procedure), FUN=sum)
+filter(Procedures, x == max(Procedures$x, na.rm=TRUE))
+
+"Task 8"
+"Which time of the day has highest frequency of visits by hour?"
+" "
+# Correct HH:MM:SS XM"
+Times <- newdf$Time
+Times <- gsub("AM",":00 AM",Times,fixed=TRUE)
+Times <- gsub("PM",":00 PM",Times,fixed=TRUE)
+for (i in 1:NROW(Times)){
+  if ((is.na(Times))==FALSE){
+  if (substr(Times,6,6) == "") {
+    Times[i] = paste(Times[i],":00 AM",sep ="")
+  }
+  }# is.na
+}
+# add new column with all NA
+newdf <- mutate(newdf,time24 = hm(Time))
+# add 12 hours to PM value and update the new mutated column
+Times2 = 0
+Times2 <- hms(Times, na.rm=TRUE)
+for (i in 1:NROW(newdf)){
+  if (grepl("P", Times[i])==TRUE){
+   Times2[i] <- hms(Times[i]) + hours(12)
+   #cntPM = cntPM + 1
+  }
+  if (grepl("A", Times[i])==TRUE){
+    Times2[i] <- hms(Times[i])
+    #cntAM = cntAM +1
+  }
+  # Update value in data fram
+  newdf$time24[i] = Times2[i]
+}
+cnt = count(newdf,hour(time24),sort=TRUE)
+cnt[1,]
+
+
+"Task 9"
+"Create a bracket of time by Morning, Afternoon, Evening, Night (6am - 12pm - Morning, 12 pm- 4 pm, Afternoon, 4 pm- 7pm, Evening, 7pm - 6 am, Night)."
+" "
+
+newdf <- mutate(newdf,TimeBracket = Time)
+#TB <- newdf$TimeBracket
+for (i in 1:NROW(newdf)){
+  if ((is.na(newdf$time24[i]))==FALSE){
+    newdf$TimeBracket[i] = paste("NA")
+  # Morning
+  if ((hour(newdf$time24[i]) >= 6)==TRUE){
+    if ((hour(newdf$time24[i]) < 12)==TRUE){
+      newdf$TimeBracket[i] = paste("Morning")
+    }
+  }
+  # Afternoon
+  if ((hour(newdf$time24[i]) >= 12)==TRUE){
+    if ((hour(newdf$time24[i]) < 16 )==TRUE){
+      newdf$TimeBracket[i] = paste("Afternoon")
+    }
+  }
+  # Evening
+  if ((hour(newdf$time24[i]) >= 16)==TRUE){
+    if ((hour(newdf$time24[i]) < 19 )==TRUE){
+      newdf$TimeBracket[i] = paste("Evening")
+    }
+  }
+  # Night
+  if ((hour(newdf$time24[i]) >= 19)==TRUE){
+    if ((hour(newdf$time24[i]) < 6 )==TRUE){
+      newdf$TimeBracket[i] = paste("Night")
+    }
+  }
+  }# master false
+}
+# Display Time Bracket
+newdf$TimeBracket
+
+
+"Task 10"
+"How many patients are repeated visitors?"
+" "
+# ID is patient
+cnt = count(newdf,id,sort=TRUE)
+count(cnt)
+
+"Task 11"
+"Give us the id of repeated visitors."
+" "
+cnt = count(newdf,id,sort=TRUE)
+filter (cnt, cnt$n > 1)
+
+"Task 12"
+"Which patients visited again for the same problem?"
+" "
+cnt = count(newdf,id,Specialty,sort=TRUE)
+cnt = filter (cnt, n > 1)
+cnt
+
+"Task 13"
+"What is the median age for Females and Males?"
+" "
+newdfMale <- filter(newdf,Sex == "M")
+"Average Male Age"
+mean(newdfMale$age,na.rm=TRUE)
+
+newdfFemale <- filter(newdf,Sex == "F")
+"Average Female Age"
+mean(newdfFemale$age,na.rm=TRUE)
+
+
+"Task 14"
+"What is the total amount in balance"
+" "
+# Remove comma before converting to numeric
+#newdf$`Amount Balance` <- mydf$`Amount Balance`
+newdf$`Amount Balance` <- gsub(",","",newdf$`Amount Balance`,fixed=TRUE)
+newdf$`Amount Balance` <- as.numeric(newdf$`Amount Balance`, na.rm=TRUE)
+"Total Amount Balance :"
+sum(newdf$`Amount Balance`,na.rm=TRUE)
+
+"Task 15"
+"How much money was made by Procedure Type 'Consultation'"
+" "
+newdf$`Amount Received ` <- as.numeric(newdf$`Amount Received `, na.rm=TRUE)
+"Revenue Generated By Procedure : Consultation "
+sum(newdf$`Amount Balance`,newdf$Procedure=='Consultation', na.rm=TRUE)
+
+"Task 16"
+"Is there a relation between Age and Total Charges paid?"
+" "
+newdf$`Total Charges` <- gsub(",","",newdf$`Total Charges`,fixed=TRUE)
+newdf$`Total Charges` <- as.numeric(newdf$`Total Charges`, na.rm=TRUE)
+plot(newdf$Age, newdf$`Total Charges`)
+#ggplot(data=newdf, aes(x=Age, y=`Total Charges`)) + theme_bw() + geom_line() #+ facet_wrap(~ variable)
+
+
+"Task 17"
+"Which Age group had highest number of visits?"
+" "
+"Age Group and Number of Visits"
+count(newdf,adult_child,sort=TRUE)
+
+
+"Task 18"
+"What is the total cost earned by Procedure Type X Ray and Scalling together?"
+" "
+newdf$`Total Charges` <- gsub(",","",newdf$`Total Charges`,fixed=TRUE)
+newdf$`Total Charges` <- as.numeric(newdf$`Total Charges`, na.rm=TRUE)
+"Total Cost Earned By Procedures 'X Ray' and 'Scalling' :"
+sum(newdf$`Total Charges`,newdf$Procedure=='X Ray', na.rm=TRUE) + sum(newdf$`Total Charges`,newdf$Procedure=='Scalling', na.rm=TRUE)
+
+"Finaly Write the clean data set"
+# load dataset
+#path2csv = "C:\\Users\\biuser\\Desktop\\DIH\\22mar17-R&P\\hospitaldata.csv"
+write.csv(newdf, file = "C:\\Users\\biuser\\Desktop\\DIH\\22mar17-R&P\\clean-data.csv")
+
+" *** END ***"