+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
#Question 1)
+library("ggplot2")
+library("lubridate")
+
##
+## Attaching package: 'lubridate'
+
## The following object is masked from 'package:base':
+##
+## date
+
library("dplyr")
+
##
+## Attaching package: 'dplyr'
+
## The following objects are masked from 'package:lubridate':
+##
+## intersect, setdiff, union
+
## 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("DataCombine")
+hospitaldata <- read.csv("hospitaldata.csv")
+names(hospitaldata)<-gsub("..","",names(hospitaldata),fixed = TRUE)
+names(hospitaldata)<-gsub(".","_",names(hospitaldata),fixed = TRUE)
+View(hospitaldata)
+
+#removing character M from age i.e 28M
+hospitaldata$Age <- as.numeric(gsub("[^0-9]",'',hospitaldata$Age))
+
+#Question 2)
+hospitaldata$Date <- as.Date(strptime(hospitaldata$Date, "%a, %B %d, %Y"))
+
+Max_day <- hospitaldata %>%
+ mutate(Day=weekdays(hospitaldata$Date),label=TRUE) %>%
+ group_by(Day) %>%
+ summarize(visits=length(Day)) %>%
+ print
+
## # A tibble: 7 × 2
+## Day visits
+## <chr> <int>
+## 1 Friday 26
+## 2 Monday 51
+## 3 Saturday 20
+## 4 Sunday 7
+## 5 Thursday 33
+## 6 Tuesday 42
+## 7 Wednesday 43
+
ggplot(Max_day,aes(x=Day,y=visits))+geom_bar(stat="identity",fill="slateblue")+ggtitle("Visits per Weekday")+labs(x="Day",y="Visits")
+
+
#Question 3)
+mean(hospitaldata$Age, na.rm = TRUE)
+
## [1] 32.57216
+
#Question 4)
+child <- filter(hospitaldata, Age > 1 & Age < 13) %>%
+ select(-(Date:Time)) %>%
+ select(-(Sex:Next_Apt)) %>%
+ count() %>%
+ print
+
## # A tibble: 1 × 1
+## n
+## <int>
+## 1 24
+
#Question 5)
+
+hospitaldata$Sex <- gsub("f","F",hospitaldata$Sex)
+hospitaldata$Sex<-gsub("\\s|-",NA,hospitaldata$Sex)
+qplot(data=hospitaldata, Sex, fill=Procedure)+ggtitle("Procedure vs Gender")+labs(x='Gender',y='Procedure')
+
+
#Question 6)
+qplot(data=hospitaldata, fill=ConsultingDoctor, as.numeric(TotalCharges))+ggtitle("Highest Salary")+labs(x='TotalCharges',y='ConsultingDoctor')
+
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
+
+
#Question 7)
+qplot(data=hospitaldata, as.numeric(TotalCharges), fill=Procedure)+ggtitle("Procedure")+labs(x='TotalCharges',y='Procedure')
+
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
+
+
#Question 8)
+hour_and_visits <- hospitaldata %>%
+ select(Time) %>%
+ mutate(Hour = hour(hm(format(strptime(hospitaldata$Time, "%I:%M %p"), "%H:%M")))) %>%
+ group_by(Hour) %>%
+ summarize(visits=length(Hour)) %>%
+ arrange(desc(visits)) %>%
+ print # printing 13 the highest hour, i.e. actaully 1 AM/PM in 12 hour format
+
## # A tibble: 16 × 2
+## Hour visits
+## <dbl> <int>
+## 1 13 29
+## 2 NA 29
+## 3 18 23
+## 4 12 20
+## 5 14 20
+## 6 15 20
+## 7 17 16
+## 8 16 15
+## 9 19 14
+## 10 11 9
+## 11 20 8
+## 12 10 6
+## 13 21 5
+## 14 22 4
+## 15 9 2
+## 16 23 2
+
#Question 9)
+
+
+#Question 10)
+visitor_repeated <- select(hospitaldata,id) %>%
+ group_by(id) %>%
+ summarise(visits=length(id)) %>%
+ arrange(desc(visits)) %>%
+ filter(visits > 1) %>%
+ print
+
## # A tibble: 37 × 2
+## id visits
+## <int> <int>
+## 1 1 12
+## 2 46 5
+## 3 122 5
+## 4 17 4
+## 5 94 4
+## 6 140 4
+## 7 45 3
+## 8 63 3
+## 9 101 3
+## 10 107 3
+## # ... with 27 more rows
+
#Question 11)
+
+visitor_repeated <- select(hospitaldata,id) %>%
+ group_by(id) %>%
+ summarise(visits=length(id)) %>%
+ arrange(desc(visits)) %>%
+ filter(visits > 1) %>%
+ print
+
## # A tibble: 37 × 2
+## id visits
+## <int> <int>
+## 1 1 12
+## 2 46 5
+## 3 122 5
+## 4 17 4
+## 5 94 4
+## 6 140 4
+## 7 45 3
+## 8 63 3
+## 9 101 3
+## 10 107 3
+## # ... with 27 more rows
+
#Question 12)
+
+hospitaldata %>%
+ count(id, Procedure) %>%
+ slice(which(n>1))%>%
+ filter(!is.na(Procedure))%>%
+ select(id,Procedure)
+
## Source: local data frame [24 x 2]
+## Groups: id [23]
+##
+## id Procedure
+## <int> <fctr>
+## 1 1 Pharmacy
+## 2 12 22 Unit Bridge
+## 3 13 Consultation
+## 4 17 Consultation
+## 5 17 RCT (4 teeth) Bridge (9 teeth)
+## 6 20 Consultation
+## 7 25 Consultation
+## 8 45 R.P.D + Crown
+## 9 46 Dressing
+## 10 63 Consultation
+## # ... with 14 more rows
+
#Question 13)
+Age_median <- hospitaldata %>%
+ select(Sex,Age) %>%
+ group_by(Sex) %>%
+ summarise(median(Age, na.rm = TRUE)) %>%
+ print
+
## # A tibble: 4 × 2
+## Sex `median(Age, na.rm = TRUE)`
+## <chr> <dbl>
+## 1 NA
+## 2 F 30
+## 3 M 29
+## 4 <NA> NA
+
#Question 14)
+hospitaldata$TotalCharges<-as.numeric(as.character(hospitaldata$TotalCharges))
+
## Warning: NAs introduced by coercion
+
sum_of_Balance <- sum(hospitaldata$TotalCharges, na.rm = TRUE)
+print(sum_of_Balance)
+
## [1] 786350
+
#Question 15)
+hospitaldata$TotalCharges<-as.numeric(as.character(hospitaldata$TotalCharges))
+consult_amount <- hospitaldata%>%
+ select(Procedure,TotalCharges) %>%
+ group_by(Procedure) %>%
+ filter(Procedure == 'Consultation') %>%
+ summarise(sum(TotalCharges,na.rm=TRUE)) %>%
+ print
+
## # A tibble: 1 × 2
+## Procedure `sum(TotalCharges, na.rm = TRUE)`
+## <fctr> <dbl>
+## 1 Consultation 83950
+
#Question 16)
+data <- hospitaldata %>%
+ select(Age,TotalCharges) %>%
+ filter(!is.na(Age),!is.na(TotalCharges))
+cor(data$Age,data$TotalCharges)
+
## [1] 0.02952065
+
#Question 17)
+Number_of_visits_by_age <- hospitaldata %>%
+ select(id,Age) %>%
+ group_by(Age) %>%
+ summarize(visits=length(Age)) %>%
+ arrange(desc(visits)) %>%
+ filter(!is.na(Age)) %>%
+ print
+
## # A tibble: 55 × 2
+## Age visits
+## <dbl> <int>
+## 1 30 20
+## 2 26 11
+## 3 40 11
+## 4 17 9
+## 5 28 9
+## 6 3 7
+## 7 45 7
+## 8 50 7
+## 9 23 6
+## 10 29 6
+## # ... with 45 more rows
+
ggplot(data=Number_of_visits_by_age,aes(x=as.numeric(Age),y=visits))+geom_bar(stat='identity',fill='slate blue')+ggtitle("Number Of Visits By Age")+labs(x='Age',y='Visits')
+
+
#Question 18)
+
+sum(hospitaldata$TotalCharges[hospitaldata$Procedure=="X Ray" | hospitaldata$Procedure=="Scalling"],na.rm = TRUE)
+
## [1] 22300
+
write.csv("~/hospitaldata.csv", row.names = FALSE)
+
## "x"
+## "~/hospitaldata.csv"
+
View(hospitaldata)
+
+
+
+
+