+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
R ASSIGNMENT 2
+
Assignment 2 of a hospital data.
+
#QUESTIONS
+#2. Which day of the week is expected to have most visits?
+print(paste("Average age of a person is",weekdays(df$Date[which(table(df$Date) == max(table(df$Date)))])))
+
## [1] "Average age of a person is Monday"
+
#3. What is the average age of patients?
+print(paste("Average age of a person is ",mean(df$Age, na.rm = T)))
+
## [1] "Average age of a person is 32.5721649484536"
+
#4. How many children were entertained? (Make a Bracket of Age from 1-12)
+total_entertained <- df %>%
+ select(Age) %>%
+ filter(Age >= 1 & Age <= 12, !is.na(.)) %>%
+ count()
+paste("Children that were entertained are", total_entertained)
+
## [1] "Children that were entertained are 24"
+
#5. Which gender type had what kind of procedure in abundance? i.e. Female visit mostly because of Gynae Problem
+df %>%
+ group_by(Sex, Procedure) %>%
+ tally(sort = T) %>%
+ filter(!is.na(Sex)) %>%
+ View()
+
+#6. Which Doctor is earning highest?
+highest_earning_doctor <- df %>%
+ group_by(ConsultingDoctor) %>%
+ summarise_each(funs(sum(TotalCharges, na.rm = T))) %>%
+ select(ConsultingDoctor, TotalCharges) %>%
+ arrange(desc(TotalCharges))
+
+paste("Highest paid doctor is",(highest_earning_doctor$ConsultingDoctor[which(highest_earning_doctor$TotalCharges == max(highest_earning_doctor$TotalCharges))]), "with pay of",max(highest_earning_doctor$TotalCharges))
+
## [1] "Highest paid doctor is Dr Alaf Khan with pay of 513050"
+
#7. Which procedure type earns more money?
+highest_earning_procedure <- df %>%
+ group_by(Procedure) %>%
+ summarise_each(funs(sum(TotalCharges, na.rm = T))) %>%
+ select(Procedure, TotalCharges) %>%
+ filter(!is.na(Procedure)) %>%
+ arrange(desc(TotalCharges))
+View(highest_earning_procedure)
+paste("Highest earning procedure is",(highest_earning_procedure$Procedure[which(highest_earning_procedure$TotalCharges == max(highest_earning_procedure$TotalCharges))]), "worth of",max(highest_earning_procedure$TotalCharges))
+
## [1] "Highest earning procedure is Orthodontics worth of 240000"
+
#8. Which time of the day has highest frequency of visits by hour?
+high_frequency_time <- df %>%
+ filter(!is.na(Date), !is.na(Time)) %>%
+ group_by(Date,Time) %>%
+ tally() %>%
+ arrange(desc(n))
+View(high_frequency_time)
+indexOfMax <- which(high_frequency_time$n == max(high_frequency_time$n))
+paste("Highest number of visits were on",weekdays(high_frequency_time$Date[indexOfMax]),"at",format(high_frequency_time$Time[indexOfMax]))
+
## [1] "Highest number of visits were on Wednesday at 02:40 PM"
+
#10. How many patients are repeated visitors?
+visitor <- df %>%
+ group_by(id) %>%
+ tally() %>%
+ filter(n > 1) %>%
+ arrange(desc(n))
+View(visitor)
+paste("Number of repeated visitors are", count(visitor))
+
## [1] "Number of repeated visitors are 37"
+
#11. Give us the id of repeated visitors.
+#Data is take from above
+View(visitor)
+print(visitor$id)
+
## [1] 1 46 122 17 94 140 45 63 101 107 109 114 132 145 4 12 13
+## [18] 20 25 40 59 64 80 88 96 97 100 112 116 118 120 130 133 149
+## [35] 150 151 153
+
#12. Which patients visited again for the same problem?
+patients_repeated <- df %>%
+ group_by(id, Procedure) %>%
+ tally() %>%
+ filter(!is.na(Procedure), n > 1) %>%
+ arrange(desc(n))
+View(patients_repeated)
+#This data show ignoring NA for procedure variable
+paste("Patient repeated most for is",(patients_repeated$Procedure[which(patients_repeated$n == max(patients_repeated$n))]))
+
## [1] "Patient repeated most for is Pharmacy"
+
#13. What is the median age for Females and Males?
+df %>%
+ group_by(Sex) %>%
+ summarise(median(Age, na.rm = T))
+
## # A tibble: 3 × 2
+## Sex `median(Age, na.rm = T)`
+## <chr> <dbl>
+## 1 F 30
+## 2 M 29
+## 3 <NA> NA
+
#14. What is the total amount in balance?
+paste("Total amount in balance is",sum(df$AmountBalance, na.rm = T))
+
## [1] "Total amount in balance is 222500"
+
#15. How much money was made by Procedure Type "Consultation"?
+df %>%
+ group_by(Procedure) %>%
+ summarise(Total_Amount = sum(TotalCharges, na.rm = T)) %>%
+ filter(Procedure %in% c("Consultation"))
+
## # A tibble: 1 × 2
+## Procedure Total_Amount
+## <chr> <dbl>
+## 1 Consultation 83950
+
#16.Is there a relation between Age and Total Charges paid
+data <- df %>%
+ filter(!is.na(Age), !is.na(TotalCharges))
+paste(cor(data$Age, data$TotalCharges),"shows positive linearly directly propotional.")
+
## [1] "0.0295206461012257 shows positive linearly directly propotional."
+
#17. Which Age group had highest number of visits?
+# HIGHEST NUMBER OF VISIT WAS BY AGE UNASIGNED 'NA' THAT IS 28 BUT IT WAS IGNORED
+highest_age_visit <- df %>%
+ group_by(Age) %>%
+ tally() %>%
+ filter(!is.na(Age)) %>%
+ arrange(desc(n))
+View(highest_age_visit)
+paste("Highest number of visit was by age group of",(highest_age_visit$Age[which(highest_age_visit$n == max(highest_age_visit$n))]), "with number of visits of",max(highest_age_visit$n))
+
## [1] "Highest number of visit was by age group of 30 with number of visits of 20"
+
#18. What is the total cost earned by Procedure Type X Ray and Scalling together?
+TotalEarning <- df %>%
+ filter(Procedure %in% c("X Ray", "Scalling")) %>%
+ summarise(Total = sum(TotalCharges, na.rm = T))
+View(TotalEarning)
+paste("Total earning by both X Ray and Scalling is worth",TotalEarning)
+
## [1] "Total earning by both X Ray and Scalling is worth 22300"
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
library(dplyr)
+
##
+## Attaching package: 'dplyr'
+
## The following objects are masked from 'package:stats':
+##
+## filter, lag
+
## The following objects are masked from 'package:base':
+##
+## intersect, setdiff, setequal, union
+
library(tidyr)
+library(lubridate)
+
##
+## Attaching package: 'lubridate'
+
## The following object is masked from 'package:base':
+##
+## date
+
#loading csv from directory
+bufferedDataFrame <- read.csv("hospitaldata.csv", strip.white = T, na.strings = c("-",""," ","\t","\n",NA), stringsAsFactors = F)
+
+#CLEANING DATA
+#converting to tbl format
+df <- tbl_df(bufferedDataFrame)
+
+#1. Removing dots(.) from column names
+names(df) <- gsub("\\.", "", names(df))
+
+#Removing character from Age Like M
+df$Age <- as.numeric(gsub("[^0-9]",'',df$Age))
+
+#Cleaning Date Column
+#Converting date factor format to Date class
+df$Date <- as.Date(strptime(df$Date, "%a, %B %d, %Y"))
+
+#Cleaning Time Column with formatting
+df$Time <- format(strptime(df$Time, format='%I:%M %p'), '%I:%M %p')
+
+#Changing case of Sex to upper case
+df$Sex <- toupper(df$Sex)
+
+#Remove Cancelled to NA as no payment is not given
+df$TotalCharges <- as.numeric(gsub("cancelled", NA, ignore.case = T,df$TotalCharges))
+
+#Remove Cancelled to NA as no consultation is not given
+df$Procedure <- (gsub("cancelled", NA, ignore.case = T,df$Procedure))
+
+#Coverting AmountBalance to numeric
+df$AmountBalance <- as.numeric(gsub(",",'',df$AmountBalance))
+class(df$AmountBalance)
+
## [1] "numeric"
+
#Exporting Clean CSV
+write.csv(df, "CleanHospitalData.csv")
+
+#View clean data
+glimpse(df)
+
## Observations: 222
+## Variables: 15
+## $ Date <date> 2017-01-01, 2017-01-02, 2017-01-02, 2017-01-...
+## $ id <int> 101, 150, 58, 75, 97, 101, 26, 149, 20, 72, 5...
+## $ Time <chr> "11:00 AM", "10:45 AM", "12:38 PM", "01:00 PM...
+## $ Age <dbl> 40, 26, 30, 40, 27, 40, 43, 28, 2, 40, 32, 28...
+## $ Sex <chr> "F", "M", "F", "M", "M", "F", "M", "F", "F", ...
+## $ ConsultingDoctor <chr> "Dr Kinza Alam", "Nursing Staff", "Dr Riffat ...
+## $ Specialty <chr> "Gynae", NA, "Psychotherapist", "Psychotherap...
+## $ Procedure <chr> "C Section", "Dressing", "Consultation", "Con...
+## $ TotalCharges <dbl> 30000, 1500, 1000, 1500, 2000, 35000, 2000, 5...
+## $ AmountReceived <int> 30000, 1500, 1000, 1500, 2000, 35000, 2000, 5...
+## $ AmountBalance <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N...
+## $ AmountReceivedBy <chr> "Mrs Shamsa", "Dr Saniya", "Mrs Shamsa", "Mrs...
+## $ AmountinHospital <int> NA, NA, 300, 450, 600, NA, NA, 500, NA, 500, ...
+## $ ReceptionistName <chr> "Hamza", "Haris", "Fiza", "Zaheer", "Haris", ...
+## $ NextApt <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, "...
+
View(df)
+
+#QUESTIONS
+#2. Which day of the week is expected to have most visits?
+print(paste("Average age of a person is",weekdays(df$Date[which(table(df$Date) == max(table(df$Date)))])))
+
## [1] "Average age of a person is Monday"
+
#3. What is the average age of patients?
+print(paste("Average age of a person is ",mean(df$Age, na.rm = T)))
+
## [1] "Average age of a person is 32.5721649484536"
+
#4. How many children were entertained? (Make a Bracket of Age from 1-12)
+total_entertained <- df %>%
+ select(Age) %>%
+ filter(Age >= 1 & Age <= 12, !is.na(.)) %>%
+ count()
+paste("Children that were entertained are", total_entertained)
+
## [1] "Children that were entertained are 24"
+
#5. Which gender type had what kind of procedure in abundance? i.e. Female visit mostly because of Gynae Problem
+df %>%
+ group_by(Sex, Procedure) %>%
+ tally(sort = T) %>%
+ filter(!is.na(Sex)) %>%
+ View()
+
+#6. Which Doctor is earning highest?
+highest_earning_doctor <- df %>%
+ group_by(ConsultingDoctor) %>%
+ summarise_each(funs(sum(TotalCharges, na.rm = T))) %>%
+ select(ConsultingDoctor, TotalCharges) %>%
+ arrange(desc(TotalCharges))
+
+paste("Highest paid doctor is",(highest_earning_doctor$ConsultingDoctor[which(highest_earning_doctor$TotalCharges == max(highest_earning_doctor$TotalCharges))]), "with pay of",max(highest_earning_doctor$TotalCharges))
+
## [1] "Highest paid doctor is Dr Alaf Khan with pay of 513050"
+
#7. Which procedure type earns more money?
+highest_earning_procedure <- df %>%
+ group_by(Procedure) %>%
+ summarise_each(funs(sum(TotalCharges, na.rm = T))) %>%
+ select(Procedure, TotalCharges) %>%
+ filter(!is.na(Procedure)) %>%
+ arrange(desc(TotalCharges))
+View(highest_earning_procedure)
+paste("Highest earning procedure is",(highest_earning_procedure$Procedure[which(highest_earning_procedure$TotalCharges == max(highest_earning_procedure$TotalCharges))]), "worth of",max(highest_earning_procedure$TotalCharges))
+
## [1] "Highest earning procedure is Orthodontics worth of 240000"
+
#8. Which time of the day has highest frequency of visits by hour?
+#It its only about number of highes frequency
+#"01:00 PM" "01:30 PM" "03:00 PM" "12:00 PM"
+high_time_repeated <- df %>%
+ filter(!is.na(Time)) %>%
+ group_by(Time) %>%
+ tally() %>%
+ arrange(desc(n))
+View(high_time_repeated)
+high_time_repeated$Time[high_time_repeated$n == max(high_time_repeated$n)]
+
## [1] "01:00 PM" "01:30 PM" "03:00 PM" "12:00 PM"
+
#If your are talking about particular day
+high_frequency_time <- df %>%
+ filter(!is.na(Date), !is.na(Time)) %>%
+ group_by(Date,Time) %>%
+ tally() %>%
+ arrange(desc(n))
+View(high_frequency_time)
+indexOfMax <- which(high_frequency_time$n == max(high_frequency_time$n))
+paste("Highest number of visits were on",weekdays(high_frequency_time$Date[indexOfMax]),"at",format(high_frequency_time$Time[indexOfMax]))
+
## [1] "Highest number of visits were on Wednesday at 02:40 PM"
+
#9. Create a bracket of time by Morning, Afternoon, Evening, Night (6am - 12pm - Morning, 12 pm- 4 pm, Afternoon, 4 pm- 7pm, Evening, 7pm - 6 am, Night).
+
+hour(strptime(df$Time, "%I:%M %p"))
+
## [1] 11 10 12 13 14 15 15 15 15 17 17 17 13 15 18 23 12 20 20 12 14 14 12
+## [24] 13 13 NA 20 NA 12 13 14 15 17 17 15 18 NA NA 15 16 16 10 14 14 11 15
+## [47] 20 16 18 21 NA 13 18 18 11 11 13 15 18 21 12 14 17 NA 11 NA NA NA 10
+## [70] 13 13 12 13 13 16 13 13 14 13 NA NA 12 15 17 NA NA NA NA 18 21 NA 13
+## [93] 13 17 17 18 17 18 18 14 14 13 13 18 12 13 13 16 20 16 16 19 19 13 13
+## [116] 16 18 12 13 14 18 20 10 12 12 14 14 14 10 9 18 19 12 16 17 18 19 12
+## [139] 11 15 19 NA 14 15 19 11 16 18 16 17 18 18 18 11 14 NA 13 14 19 22 13
+## [162] 18 19 22 NA 15 16 17 13 13 14 15 21 15 16 11 12 NA 22 12 NA 15 20 17
+## [185] 18 NA 19 19 12 15 19 NA 11 12 13 16 17 14 NA 13 17 19 NA 15 NA 19 14
+## [208] 21 20 22 16 18 12 19 12 9 NA NA 15 18 10 23
+
df %>%
+ mutate(TimeBracket = ifelse(hour(strptime(df$Time, "%I:%M %p")) >= 6 & hour(strptime(df$Time, "%I:%M %p")) <= 12, "Morning",
+ ifelse(hour(strptime(df$Time, "%I:%M %p")) > 12 & hour(strptime(df$Time, "%I:%M %p")) <= 16, "Afternoon",
+ ifelse(hour(strptime(df$Time, "%I:%M %p")) > 16 & hour(strptime(df$Time, "%I:%M %p")) <= 19, "Evening",
+ ifelse(hour(strptime(df$Time, "%I:%M %p")) > 19, "Night", NA))))) %>%
+ View()
+
+
+
+#10. How many patients are repeated visitors?
+visitor <- df %>%
+ group_by(id) %>%
+ tally() %>%
+ filter(n > 1) %>%
+ arrange(desc(n))
+View(visitor)
+paste("Number of repeated visitors are", count(visitor))
+
## [1] "Number of repeated visitors are 37"
+
#11. Give us the id of repeated visitors.
+#Data is take from above
+View(visitor)
+print(visitor$id)
+
## [1] 1 46 122 17 94 140 45 63 101 107 109 114 132 145 4 12 13
+## [18] 20 25 40 59 64 80 88 96 97 100 112 116 118 120 130 133 149
+## [35] 150 151 153
+
#12. Which patients visited again for the same problem?
+patients_repeated <- df %>%
+ group_by(id, Procedure) %>%
+ tally() %>%
+ filter(!is.na(Procedure), n > 1) %>%
+ arrange(desc(n))
+View(patients_repeated)
+#This data show ignoring NA for procedure variable
+paste("Patient repeated most for is",(patients_repeated$Procedure[which(patients_repeated$n == max(patients_repeated$n))]))
+
## [1] "Patient repeated most for is Pharmacy"
+
#13. What is the median age for Females and Males?
+df %>%
+ group_by(Sex) %>%
+ summarise(median(Age, na.rm = T))
+
## # A tibble: 3 × 2
+## Sex `median(Age, na.rm = T)`
+## <chr> <dbl>
+## 1 F 30
+## 2 M 29
+## 3 <NA> NA
+
#14. What is the total amount in balance?
+paste("Total amount in balance is",sum(df$AmountBalance, na.rm = T))
+
## [1] "Total amount in balance is 222500"
+
#15. How much money was made by Procedure Type "Consultation"?
+df %>%
+ group_by(Procedure) %>%
+ summarise(Total_Amount = sum(TotalCharges, na.rm = T)) %>%
+ filter(Procedure %in% c("Consultation"))
+
## # A tibble: 1 × 2
+## Procedure Total_Amount
+## <chr> <dbl>
+## 1 Consultation 83950
+
#16.Is there a relation between Age and Total Charges paid
+data <- df %>%
+ filter(!is.na(Age), !is.na(TotalCharges))
+paste(cor(data$Age, data$TotalCharges),"shows positive linearly directly propotional.")
+
## [1] "0.0295206461012257 shows positive linearly directly propotional."
+
#17. Which Age group had highest number of visits?
+# HIGHEST NUMBER OF VISIT WAS BY AGE UNASIGNED 'NA' THAT IS 28 BUT IT WAS IGNORED
+highest_age_visit <- df %>%
+ group_by(Age) %>%
+ tally() %>%
+ filter(!is.na(Age)) %>%
+ arrange(desc(n))
+View(highest_age_visit)
+paste("Highest number of visit was by age group of",(highest_age_visit$Age[which(highest_age_visit$n == max(highest_age_visit$n))]), "with number of visits of",max(highest_age_visit$n))
+
## [1] "Highest number of visit was by age group of 30 with number of visits of 20"
+
#18. What is the total cost earned by Procedure Type X Ray and Scalling together?
+TotalEarning <- df %>%
+ filter(Procedure %in% c("X Ray", "Scalling")) %>%
+ summarise(Total = sum(TotalCharges, na.rm = T))
+View(TotalEarning)
+paste("Total earning by both X Ray and Scalling is worth",TotalEarning)
+
## [1] "Total earning by both X Ray and Scalling is worth 22300"
+
+
+
+
+