Bike_Sharing_Analysis.Rmd

---
title: "Cyclistic_bikes_questions"
author: "ysnmslk"
date: "2024-04-27"
output:
  word_document: default
  html_document: default
  pdf_document: default
editor_options:
  markdown:
    wrap: 72
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

# Question of Analysis

#### 1 How do annual members and casual riders use Cyclistic bikes differently?

#### 2. Why would casual riders buy Cyclistic annual memberships?

#### 3. How can Cyclistic use digital media to influence casual riders to become members?

######################################################################################## 
######################################################################################## 
######################################################################################## 

#### Kütüphanelerin yüklenmesi

#### Load Packages

install.packages("tidyverse")

library(readr)
library(tidyverse)
library(dplyr)
library(lubridate)
library(skimr)
library(janitor)
library(ggplot2)

######################################################################################## 
######################################################################################## 
######################################################################################## 

## Verisetlerini Okuma

## Read Datasets

td2023_1 <- read_csv("202301-divvy-tripdata.csv") # File name or full path of the file
td2023_2 <- read_csv("202302-divvy-tripdata.csv") # File name or full path of the file
td2023_3 <- read_csv("202303-divvy-tripdata.csv") # File name or full path of the file
td2023_4 <- read_csv("202304-divvy-tripdata.csv") # File name or full path of the file
td2023_5 <- read_csv("202305-divvy-tripdata.csv") # File name or full path of the file
td2023_6 <- read_csv("202306-divvy-tripdata.csv") # File name or full path of the file
td2023_7 <- read_csv("202307-divvy-tripdata.csv") # File name or full path of the file
td2023_8 <- read_csv("202308-divvy-tripdata.csv") # File name or full path of the file
td2023_9 <- read_csv("202309-divvy-tripdata.csv") # File name or full path of the file
td2023_10 <- read_csv("202310-divvy-tripdata.csv") # File name or full path of the file
td2023_11 <- read_csv("202311-divvy-tripdata.csv") # File name or full path of the file
td2023_12 <- read_csv("202312-divvy-tripdata.csv") # File name or full path of the file



View(td2023_1)
View(td2023_2)
View(td2023_3)
View(td2023_4)
View(td2023_5)
View(td2023_6)
View(td2023_7)
View(td2023_8)
View(td2023_9)
View(td2023_10)
View(td2023_11)
View(td2023_12)


## Understanding Dataset



## The skim without charts function gives us a pretty comprehensive summary of a dataset.

skim_without_charts(td2023_1) 
skim_without_charts(td2023_2)

summary(td2023_12) 
summary(td2023_11)

glimpse(td2023_10) 
glimpse(td2023_9)

colnames(td2023_8)
colnames(td2023_7)

str(td2023_5)
str(td2023_6)


colnames(td2023_7)

#### [1] "ride_id"            "rideable_type"      "started_at"         "ended_at"           "start_station_name" "start_station_id"   "end_station_name"  
####  [8] "end_station_id"     "start_lat"          "start_lng"          "end_lat"            "end_lng"            "member_casual"    

#### After check all datasets and be sure all of same columns we can bind all of them

data_2023 <- rbind(td2023_1,td2023_2,td2023_3,td2023_4,td2023_5,td2023_6,td2023_7,td2023_8,td2023_9,td2023_10,td2023_11,td2023_12)

str(data_2023)
dim(data_2023)

#### Rename of columnnames

  (data_2023 <- rename(data_2023, 
          bikeid = ride_id , 
          biketype =rideable_type,
          start_time= started_at, 
          end_time = ended_at,
          from_station_name= start_station_name ,
          from_station_id =start_station_id ,
          to_station_name = end_station_name ,
          to_station_id =end_station_id, 
          usertype = member_casual ))

head(data_2023)

#It is always good to backup raw data before data cleaning.
#Write as a csv file

write_csv(data_2023, "total_data_2023.csv") 

head(data_2023,12) #First 12 row we can see

## Data Cleaning

#### Preprocessing of the data is important before analysis, so null values have to be checked and removed.

str(data_2023)

View(data_2023)

glimpse(data_2023)

dim(data_2023)

sum(is.null(data_2023)) #be sure that get rid of null datas. 


#In this step we generate new colums that dates split to see under part of date when we analyze

data_2023$date <- as.Date(data_2023$start_time)
data_2023$day <- format(as.Date(data_2023$date),"%d")
data_2023$month <- format(as.Date(data_2023$date),"%m")
data_2023$year <- format(as.Date(data_2023$date),"%Y") #to get the year as separate column
data_2023$day_of_week <- format(as.Date(data_2023$date),"%A")

colnames(data_2023)
head(data_2023)
View(data_2023)

# After adding extra date columns, we save again. because we may want to continue the analysis with different tools.
write_csv(data_2023, "extra_columns_total_data_2023.csv") 


# Remove lat, long, birthyear, and gender fields as this data was dropped beginning in 2020

data_2023 <- data_2023 %>% 
select(-c(start_lat,start_lng, end_lat,end_lng, from_station_id, from_station_name, to_station_name, to_station_id  ))

colnames(data_2023)


######################################################################################## 
######################################################################################## 
######################################################################################## 

#### STEP 4: Clean up and prepare for data analysis

colnames(data_2023) #List of column names 
nrow(data_2023) #How many rows are in data frame? 
dim(data_2023) #Dimensions of the data frame?
head(data_2023) #See the first 6 rows of data frame. tail(div_2019_2)
str(data_2023) #See list of columns and data types (numeric, character,etc) 
summary(data_2023) #Statistical summary of data. Mainly for numerics


# To learn each of column null data

colSums(is.na(data_2023))

  # calculate the trip time
  * The resulted ride_time is given in the 'Seconds' so we have converted it into 'Minutes' by dividing it by 60.
?difftime()

data_2023$tripduration <- difftime(data_2023$end_time, data_2023$start_time, units= "mins") #Second way to find same things

summary(data_2023$tripduration)
head(data_2023)


# To know the unique values from 'usertype '
unique(data_2023$usertype )


# To know the unique values from 'biketype '
unique(data_2023$usertype )



# Convert "tripduration" from Factor to numeric so we can run calculations on the data

is.factor(data_2023$tripduration) 
data_2023$tripduration <-as.numeric(as.character(data_2023$tripduration))

is.factor(data_2023$tripduration) 
is.numeric(data_2023$tripduration)


# Removed rows which had negative tripduration but I dont want to so under 10 minutes trips effect analyze 
data_2023 <- data_2023 %>%
  filter(tripduration > 10)
  


######################################################################################## 
######################################################################################## 
######################################################################################## 

#### STEP 5: CONDUCT DESCRIPTIVE ANALYSIS

# to save original datas we analyze copy to datas a new file

trip_datas <- data_2023


colnames(trip_datas)



table(trip_datas$usertype) //Table of usertype
table(trip_datas$biketype) //Table of biketype

bikeofusertype <- table(trip_datas$usertype, trip_datas$biketype ) #which user choose which bike


?write.table()
write.table(bikeofusertype, file = "Usertype_ofbik_chose.csv", sep = ",", quote = FALSE, row.names = TRUE )

View(trip_datas)

#### Descriptive analysis on ride_length (all figures in seconds)



summary(trip_datas$tripduration)

# You can condense the four lines above to one line using summary() on the specific attribute

summary(trip_datas$tripduration)

# Compare members and casual users

aggregate(trip_datas$tripduration~trip_datas$usertype,FUN=max)

min_trip_duration <- aggregate(trip_datas$tripduration~trip_datas$usertype,FUN=min)
average_trip_duration <-aggregate(trip_datas$tripduration~trip_datas$usertype,FUN=mean)
median_trip_duration <-aggregate(trip_datas$tripduration~trip_datas$usertype,FUN=median)
max_trip_duration <-aggregate(trip_datas$tripduration~trip_datas$usertype,FUN=max)

trip_duration_datas <- rbind(min_trip_duration, average_trip_duration,median_trip_duration,max_trip_duration )

write.table(trip_duration_datas, file = "trip_duration_datas.csv", sep = ",", quote = FALSE,  row.names = TRUE )

summary(trip_datas$tripduration)


# Sort days of the week # Notice that the days of the week are out of order. Let's fix that.

trip_datas$day_of_week <- ordered(trip_datas$day_of_week, levels=c("Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"))



## Comparing by each day of week for member vs casual
datasof_usertype_days <- trip_datas %>%
        group_by(usertype, day_of_week) %>%
        summarise(
                average_ride_duration = mean(tripduration),
                max = max(tripduration),
                min = min(tripduration)) %>%
        arrange(usertype, day_of_week)

View(datasof_usertype_days)
  
  write.table(datasof_usertype_days, file = "datasof_usertype_day.csv", sep = ",", quote = FALSE,  row.names = TRUE )

ggplot(datasof_usertype_days, aes(x = day_of_week, y = average_ride_duration, fill = usertype, colour = usertype)) + 
geom_bar(position = "dodge", stat="identity")
  

######################################################################################## 
######################################################################################## 
######################################################################################## 

View(trip_datas)
colnames(trip_datas)

## It's time to Visualization to share MOrena 
#### Plot an bar_chart to compare the number of rides visually

#Total number of rides for the year 2023
ggplot(data = trip_datas)+
  geom_bar(mapping = aes(x=usertype,  fill=usertype),show.legend = FALSE,width = 0.8)+
  labs(y="total",title = "Total number of rides for the year 2023")
    ggsave("Total number of rides for the year 2023.png")
 
 
 
 ?pie() 
# Create a data frame of member_casual_count_summary
type_of_users_summary<- trip_datas %>% 
                group_by(usertype) %>% 
                summarise(count=n())
                              
# Create a labels for pie chart
pie_labels <- paste0(round(100*type_of_users_summary$count/sum(type_of_users_summary$count),2),"%")

pie(type_of_users_summary$count, labels = as.character(pie_labels), lty=2, col =  c("lightblue", "pink"), main = "% of ride by customer type")


# __2) Total rides taken per month for each type of customers__  


# Select the required columns & reshape the data
monthly_ride_count <- trip_datas %>% 
                      group_by(month,usertype) %>% 
                      summarise(count_of_ride =n())

# Plot the line chart 
ggplot(monthly_ride_count,aes(x=month,y=count_of_ride,group=usertype,xlim(0,400000)))+
  geom_point(aes(color=usertype),size=1.5)+
  geom_line(aes(color=usertype),size=1)+
  labs(x="Month",y="Count_of_rides",title = "Total No. of rides per month")


## Monthly ride count difference between member and casual riders

# Created an subset of data frame

monthly_ride_count_2 <- monthly_ride_count %>%
          pivot_wider(names_from = usertype,values_from = count_of_ride) %>% #use pivot_mutate(ride_count_diff = member-casual)
          mutate(ride_count_diff = member-casual)
print(monthly_ride_count_2)
      
## Average of monthly ride count difference between member and casual riders
      
 mean(monthly_ride_count_2$ride_count_diff)     
      
      
# __3) Total rides taken each day in a week for each type of customers__
  
 colnames(trip_datas) 
 day_wise_ride_count <- trip_datas %>% 
                      group_by(day_of_week,usertype) %>% 
                      summarise(count_of_ride =n()) %>% 
                      pivot_wider(names_from = usertype,values_from = count_of_ride) %>% 
                      mutate(total_rides = casual+member) %>% 
                      mutate(casual_percentage= (casual/total_rides)*100) %>% 
                      mutate(member_percentage= (member/total_rides)*100)
print(day_wise_ride_count) 
  
  
day_wise_ride_count_2 <- trip_datas %>% 
                      group_by(day_of_week,usertype) %>% 
                      summarise(count_of_ride =n())
ggplot(data = day_wise_ride_count_2)+
  geom_col(mapping=aes(x=day_of_week,y=count_of_ride,fill=usertype))+
  labs(x="day_of_week",y="Count_of_rides",title = "Day wise total ride count")  
  
  
  
  View(trip_datas)
# __4) Types of bikes per type of customers__
  
bike_type_count <- trip_datas %>% 
                   group_by(biketype) %>% 
                   summarise(ride_count=n()) %>% 
                   mutate(ride_count_percentage = round(100*ride_count/sum(ride_count),1))
print(bike_type_count)
  
 pie(bike_type_count$ride_count,labels = paste0(bike_type_count$ride_count_percentage,"%"),col = rainbow(length(bike_type_count$ride_count)))
legend(bike_type_count$rideable_type,cex = 0.8,fill = rainbow(length(bike_type_count$ride_count))) 
  
  
bike_type_count_2 <- trip_datas %>% 
                   group_by(biketype,usertype) %>%
                   summarise(ride_count=n()) %>% 
                   pivot_wider(names_from = usertype,values_from = ride_count) %>% 
                   mutate(total_ride_count = sum(casual,member,na.rm=TRUE)) %>% 
                   mutate(casual_percentage = round(100*(casual/total_ride_count),2)) %>% 
                   mutate(member_percentage = round(100*(member/total_ride_count),2))
print(bike_type_count_2)
  
  
  
  
bike_type_count_3 <- trip_datas %>% 
                   group_by(biketype,usertype) %>%
                   summarise(ride_count=n())
ggplot(data = bike_type_count_3)+
  geom_col(mapping=aes(x=biketype,y=ride_count,fill=usertype))+
  labs(x="biketype",y="total_rides",title = "Total number of rides/rideable type")
  
  
# __V) Share & Act__

I will skip the following steps __Share__ and __Act__ because, the analysis is a personal project.
I will go directly to findings and recommendations.   
  
# __VI) Findings and Recommendations__  

## __1) How do annual members and casual riders use Cyclistic bikes differently?__