Description :
Airbnb is a platform that allows house and apartment owners to rent their properties to guests for short-term stays.
- Since 2011, hosts have been using Airbnb. This dataset describes the listing activity and metrics in NYC for 2023.
NYC
PHASE 1 : Ask
About the company:
Airbnb, Inc is an American San Francisco-based company operating an online marketplace for short- and long-term homestays and experiences.
The company was founded in 2008 by Brian Chesky, Nathan Blecharczyk, and Joe Gebbia.
Since it was founded in 2008, Airbnb has become one of the most successful and valuable start-ups in the world and has significantly impacted the HORECA (hotel, restaurant, and catering) industry.
Content:
- This Dataset includes information about hosts, geographical availability, and necessary metrics to draw data-driven decision-making.
Key task :
- Identified the business task.
- Considered key stakeholders.
Deliverable :
- To gain insights from Data to solve business problem.
PHASE 2 : Prepare
- Before we begin, there are few key points that are wrapped below as these are the vital steps I’ll be following to ensure its completion:
This is a Public Dataset is part of Airbnb, and the original source can be found on InsideAirbnb.
Downloaded the data and stored it on my Google Drive
The data has been made available by Inside Airbnb with No Copyright CC0: Public Domain
Using the ROCCC System to determine the credibility and integrity of the data.
Reliability: This data is reliable. This public dataset is a subset of Airbnb data and is made available for public use.
Originality: This is Original subset dataset.
Comprehensiveness: This data is comprehensive. It provides comprehensive information about Airbnb listings, hosts, and various metrics for analysis and research purposes.
Current: A recent dataset, which is current.
Cited: Inside Airbnb created the dataset, made it Public Dataset so this is Credible
- Therefore, the data is not Biased And have full credibility for the same reason. It meets ROCCC System since it’s reliable, original, comprehensive, current and cited.
Key task :
- Downloaded data and stored it appropriately.
- Identified how it’s organized.
- Determined the credibility of the data.
- Considered key stakeholders.
Deliverable :
- This dataset describes the listing activities for NYC from 2011 - 2023.
PHASE 3 : Process
- I am using R since it is an Advanced Language that performs various complex Statistical computations, Analysis, Mining.. Therefore, it is widely used by Data Scientists. Hence, I chose R.
- Dependencies :
# install.packages("tidyverse")
# install.packages("dplyr")
# install.packages("skimr")
# install.packages("mice")
# install.packages("randomForest")
# install.packages("corrplot")
# install.packages("ggcorrplot")- Libraries :
library(tidyverse)
library(dplyr)
library(skimr)
library(mice)
library(randomForest)
library(corrplot)
library(ggcorrplot)- Working Directory :
- Data Collection :
Data Wrangling :
- Ensured Data’s integrity.
- Ensured column(s) name consistent.
- A quick summary before proceeding :
## id name host_id host_name
## Min. :2.595e+03 Length:42931 Min. : 1678 Length:42931
## 1st Qu.:1.940e+07 Class :character 1st Qu.: 16085328 Class :character
## Median :4.337e+07 Mode :character Median : 74338125 Mode :character
## Mean :2.223e+17 Mean :151601209
## 3rd Qu.:6.305e+17 3rd Qu.:268069240
## Max. :8.405e+17 Max. :503872891
##
## neighbourhood_group neighbourhood latitude longitude
## Length:42931 Length:42931 Min. :40.50 Min. :-74.25
## Class :character Class :character 1st Qu.:40.69 1st Qu.:-73.98
## Mode :character Mode :character Median :40.72 Median :-73.95
## Mean :40.73 Mean :-73.94
## 3rd Qu.:40.76 3rd Qu.:-73.92
## Max. :40.91 Max. :-73.71
##
## room_type price minimum_nights number_of_reviews
## Length:42931 Min. : 0.0 Min. : 1.00 Min. : 0.00
## Class :character 1st Qu.: 75.0 1st Qu.: 2.00 1st Qu.: 1.00
## Mode :character Median : 125.0 Median : 7.00 Median : 5.00
## Mean : 200.3 Mean : 18.11 Mean : 25.86
## 3rd Qu.: 200.0 3rd Qu.: 30.00 3rd Qu.: 24.00
## Max. :99000.0 Max. :1250.00 Max. :1842.00
##
## last_review reviews_per_month calculated_host_listings_count
## Length:42931 Min. : 0.010 Min. : 1.00
## Class :character 1st Qu.: 0.140 1st Qu.: 1.00
## Mode :character Median : 0.520 Median : 1.00
## Mean : 1.169 Mean : 24.05
## 3rd Qu.: 1.670 3rd Qu.: 4.00
## Max. :86.610 Max. :526.00
## NA's :10304
## availability_365 number_of_reviews_ltm license
## Min. : 0.0 Min. : 0.000 Length:42931
## 1st Qu.: 0.0 1st Qu.: 0.000 Class :character
## Median : 89.0 Median : 0.000 Mode :character
## Mean :140.3 Mean : 7.737
## 3rd Qu.:289.0 3rd Qu.: 7.000
## Max. :365.0 Max. :1093.000
## - Renaming few columns to understand the data more easily :
nyc_df <- nyc_list %>%
rename(list_id = id,
listing_name = name,
area = neighbourhood_group,
geo_location = neighbourhood,
host_list_count = calculated_host_listings_count,
reviews_per_year = number_of_reviews_ltm,
reviews_per_month_pct = reviews_per_month) %>%
select(-license)
nyc_df - Removed ‘license’ column since it had only one value for one host id
- missing & empty values in dataset
| Name | nyc_df |
| Number of rows | 42931 |
| Number of columns | 17 |
| _______________________ | |
| Column type frequency: | |
| character | 6 |
| numeric | 11 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| listing_name | 0 | 1 | 0 | 249 | 10 | 41410 | 0 |
| host_name | 0 | 1 | 0 | 35 | 5 | 9832 | 0 |
| area | 0 | 1 | 5 | 13 | 0 | 5 | 0 |
| geo_location | 0 | 1 | 4 | 25 | 0 | 223 | 0 |
| room_type | 0 | 1 | 10 | 15 | 0 | 4 | 0 |
| last_review | 0 | 1 | 0 | 10 | 10304 | 2796 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 |
|---|---|---|---|---|---|---|---|---|---|
| list_id | 0 | 1.00 | 2.222772e+17 | 3.344213e+17 | 2595.00 | 19404736.00 | 43374815.00 | 6.305016e+17 | 8.404660e+17 |
| host_id | 0 | 1.00 | 1.516012e+08 | 1.621301e+08 | 1678.00 | 16085328.00 | 74338125.00 | 2.680692e+08 | 5.038729e+08 |
| latitude | 0 | 1.00 | 4.073000e+01 | 6.000000e-02 | 40.50 | 40.69 | 40.72 | 4.076000e+01 | 4.091000e+01 |
| longitude | 0 | 1.00 | -7.394000e+01 | 6.000000e-02 | -74.25 | -73.98 | -73.95 | -7.392000e+01 | -7.371000e+01 |
| price | 0 | 1.00 | 2.003100e+02 | 8.950800e+02 | 0.00 | 75.00 | 125.00 | 2.000000e+02 | 9.900000e+04 |
| minimum_nights | 0 | 1.00 | 1.811000e+01 | 2.746000e+01 | 1.00 | 2.00 | 7.00 | 3.000000e+01 | 1.250000e+03 |
| number_of_reviews | 0 | 1.00 | 2.586000e+01 | 5.662000e+01 | 0.00 | 1.00 | 5.00 | 2.400000e+01 | 1.842000e+03 |
| reviews_per_month_pct | 10304 | 0.76 | 1.170000e+00 | 1.790000e+00 | 0.01 | 0.14 | 0.52 | 1.670000e+00 | 8.661000e+01 |
| host_list_count | 0 | 1.00 | 2.405000e+01 | 8.087000e+01 | 1.00 | 1.00 | 1.00 | 4.000000e+00 | 5.260000e+02 |
| availability_365 | 0 | 1.00 | 1.402600e+02 | 1.420000e+02 | 0.00 | 0.00 | 89.00 | 2.890000e+02 | 3.650000e+02 |
| reviews_per_year | 0 | 1.00 | 7.740000e+00 | 1.829000e+01 | 0.00 | 0.00 | 0.00 | 7.000000e+00 | 1.093000e+03 |
- Changed datatype for last_review :
- Replacing missing value with NA :
- Data Imputation needed since everything looks good but these two
columns :
- changed datatypes before using mice ;
nyc_df$reviews_per_year <- as.integer(nyc_df$reviews_per_year)
nyc_df$last_review <- as.integer(nyc_df$last_review)
nyc_df <- as.data.frame(nyc_df)
skim_without_charts(nyc_df)| Name | nyc_df |
| Number of rows | 42931 |
| Number of columns | 17 |
| _______________________ | |
| Column type frequency: | |
| character | 5 |
| numeric | 12 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| listing_name | 10 | 1 | 1 | 249 | 0 | 41409 | 0 |
| host_name | 5 | 1 | 1 | 35 | 0 | 9831 | 0 |
| area | 0 | 1 | 5 | 13 | 0 | 5 | 0 |
| geo_location | 0 | 1 | 4 | 25 | 0 | 223 | 0 |
| room_type | 0 | 1 | 10 | 15 | 0 | 4 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 |
|---|---|---|---|---|---|---|---|---|---|
| list_id | 0 | 1.00 | 2.222772e+17 | 3.344213e+17 | 2595.00 | 19404736.00 | 43374815.00 | 6.305016e+17 | 8.404660e+17 |
| host_id | 0 | 1.00 | 1.516012e+08 | 1.621301e+08 | 1678.00 | 16085328.00 | 74338125.00 | 2.680692e+08 | 5.038729e+08 |
| latitude | 0 | 1.00 | 4.073000e+01 | 6.000000e-02 | 40.50 | 40.69 | 40.72 | 4.076000e+01 | 4.091000e+01 |
| longitude | 0 | 1.00 | -7.394000e+01 | 6.000000e-02 | -74.25 | -73.98 | -73.95 | -7.392000e+01 | -7.371000e+01 |
| price | 0 | 1.00 | 2.003100e+02 | 8.950800e+02 | 0.00 | 75.00 | 125.00 | 2.000000e+02 | 9.900000e+04 |
| minimum_nights | 0 | 1.00 | 1.811000e+01 | 2.746000e+01 | 1.00 | 2.00 | 7.00 | 3.000000e+01 | 1.250000e+03 |
| number_of_reviews | 0 | 1.00 | 2.586000e+01 | 5.662000e+01 | 0.00 | 1.00 | 5.00 | 2.400000e+01 | 1.842000e+03 |
| last_review | 10304 | 0.76 | 1.885591e+04 | 8.011800e+02 | 15106.00 | 18331.00 | 19319.00 | 1.938900e+04 | 1.942200e+04 |
| reviews_per_month_pct | 10304 | 0.76 | 1.170000e+00 | 1.790000e+00 | 0.01 | 0.14 | 0.52 | 1.670000e+00 | 8.661000e+01 |
| host_list_count | 0 | 1.00 | 2.405000e+01 | 8.087000e+01 | 1.00 | 1.00 | 1.00 | 4.000000e+00 | 5.260000e+02 |
| availability_365 | 0 | 1.00 | 1.402600e+02 | 1.420000e+02 | 0.00 | 0.00 | 89.00 | 2.890000e+02 | 3.650000e+02 |
| reviews_per_year | 0 | 1.00 | 7.740000e+00 | 1.829000e+01 | 0.00 | 0.00 | 0.00 | 7.000000e+00 | 1.093000e+03 |
- Classification and Regression Using Mice, method
cart.
- For last_review & reviews_per_month_pct.
##
## iter imp variable
## 1 1 last_review reviews_per_month_pct
## 1 2 last_review reviews_per_month_pct
## 1 3 last_review reviews_per_month_pct
## 1 4 last_review reviews_per_month_pct
## 1 5 last_review reviews_per_month_pct
## 1 6 last_review reviews_per_month_pct
## 1 7 last_review reviews_per_month_pct
## 1 8 last_review reviews_per_month_pct
## 1 9 last_review reviews_per_month_pct
## 1 10 last_review reviews_per_month_pct
## 2 1 last_review reviews_per_month_pct
## 2 2 last_review reviews_per_month_pct
## 2 3 last_review reviews_per_month_pct
## 2 4 last_review reviews_per_month_pct
## 2 5 last_review reviews_per_month_pct
## 2 6 last_review reviews_per_month_pct
## 2 7 last_review reviews_per_month_pct
## 2 8 last_review reviews_per_month_pct
## 2 9 last_review reviews_per_month_pct
## 2 10 last_review reviews_per_month_pct
## 3 1 last_review reviews_per_month_pct
## 3 2 last_review reviews_per_month_pct
## 3 3 last_review reviews_per_month_pct
## 3 4 last_review reviews_per_month_pct
## 3 5 last_review reviews_per_month_pct
## 3 6 last_review reviews_per_month_pct
## 3 7 last_review reviews_per_month_pct
## 3 8 last_review reviews_per_month_pct
## 3 9 last_review reviews_per_month_pct
## 3 10 last_review reviews_per_month_pct
## 4 1 last_review reviews_per_month_pct
## 4 2 last_review reviews_per_month_pct
## 4 3 last_review reviews_per_month_pct
## 4 4 last_review reviews_per_month_pct
## 4 5 last_review reviews_per_month_pct
## 4 6 last_review reviews_per_month_pct
## 4 7 last_review reviews_per_month_pct
## 4 8 last_review reviews_per_month_pct
## 4 9 last_review reviews_per_month_pct
## 4 10 last_review reviews_per_month_pct
## 5 1 last_review reviews_per_month_pct
## 5 2 last_review reviews_per_month_pct
## 5 3 last_review reviews_per_month_pct
## 5 4 last_review reviews_per_month_pct
## 5 5 last_review reviews_per_month_pct
## 5 6 last_review reviews_per_month_pct
## 5 7 last_review reviews_per_month_pct
## 5 8 last_review reviews_per_month_pct
## 5 9 last_review reviews_per_month_pct
## 5 10 last_review reviews_per_month_pct## Warning: Number of logged events: 5- Formatting Data Types & Mutating New Column :
nyc_df$host_id <- as.numeric(nyc_df$host_id)
nyc_df$price <- as.numeric(nyc_df$price)
nyc_df$minimum_nights <- as.numeric(nyc_df$minimum_nights)
nyc_df$number_of_reviews <- as.numeric(nyc_df$number_of_reviews)
nyc_df$last_review <- as.Date(nyc_df$last_review)
nyc_df$year <- format(as.Date(nyc_df$last_review), "%Y") # mutated year- Replicating to new dataframe :
| Name | test_com |
| Number of rows | 42931 |
| Number of columns | 18 |
| _______________________ | |
| Column type frequency: | |
| character | 6 |
| Date | 1 |
| numeric | 11 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| listing_name | 10 | 1 | 1 | 249 | 0 | 41409 | 0 |
| host_name | 5 | 1 | 1 | 35 | 0 | 9831 | 0 |
| area | 0 | 1 | 5 | 13 | 0 | 5 | 0 |
| geo_location | 0 | 1 | 4 | 25 | 0 | 223 | 0 |
| room_type | 0 | 1 | 10 | 15 | 0 | 4 | 0 |
| year | 0 | 1 | 4 | 4 | 0 | 13 | 0 |
Variable type: Date
| skim_variable | n_missing | complete_rate | min | max | median | n_unique |
|---|---|---|---|---|---|---|
| last_review | 0 | 1 | 2011-05-12 | 2023-03-06 | 2022-03-06 | 2795 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 |
|---|---|---|---|---|---|---|---|---|---|
| list_id | 0 | 1 | 2.222772e+17 | 3.344213e+17 | 2595.00 | 19404736.00 | 43374815.00 | 6.305016e+17 | 8.404660e+17 |
| host_id | 0 | 1 | 1.516012e+08 | 1.621301e+08 | 1678.00 | 16085328.00 | 74338125.00 | 2.680692e+08 | 5.038729e+08 |
| latitude | 0 | 1 | 4.073000e+01 | 6.000000e-02 | 40.50 | 40.69 | 40.72 | 4.076000e+01 | 4.091000e+01 |
| longitude | 0 | 1 | -7.394000e+01 | 6.000000e-02 | -74.25 | -73.98 | -73.95 | -7.392000e+01 | -7.371000e+01 |
| price | 0 | 1 | 2.003100e+02 | 8.950800e+02 | 0.00 | 75.00 | 125.00 | 2.000000e+02 | 9.900000e+04 |
| minimum_nights | 0 | 1 | 1.811000e+01 | 2.746000e+01 | 1.00 | 2.00 | 7.00 | 3.000000e+01 | 1.250000e+03 |
| number_of_reviews | 0 | 1 | 2.586000e+01 | 5.662000e+01 | 0.00 | 1.00 | 5.00 | 2.400000e+01 | 1.842000e+03 |
| reviews_per_month_pct | 0 | 1 | 9.100000e-01 | 1.630000e+00 | 0.01 | 0.08 | 0.25 | 1.170000e+00 | 8.661000e+01 |
| host_list_count | 0 | 1 | 2.405000e+01 | 8.087000e+01 | 1.00 | 1.00 | 1.00 | 4.000000e+00 | 5.260000e+02 |
| availability_365 | 0 | 1 | 1.402600e+02 | 1.420000e+02 | 0.00 | 0.00 | 89.00 | 2.890000e+02 | 3.650000e+02 |
| reviews_per_year | 0 | 1 | 7.740000e+00 | 1.829000e+01 | 0.00 | 0.00 | 0.00 | 7.000000e+00 | 1.093000e+03 |
- Done Process part.
Key task :
- Imputed by classification and regression trees.
- Manipulated empty data.
- Imputed missing data.
- Renamed columns for better data understanding.
- Converted Data Types.
- Created fresh data frame.
- listing_name has 10 missing values & host_name has 5 missing values.
Deliverable :
- The dataset is ready for Analysis.
PHASE 4 : Analysis
Descriptive analyses are being used to summarize and explore the behavior of the data.
- Using statistical techniques to understand the pattern.
Total Hosts
- There are 27455 unique host_id
- 42931 Total columns
## [1] 27455## [1] 42931 18- Price
- As minimum price listed $0
- As average price $200
- As maximum price 99000
- As median price 125
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0 75.0 125.0 200.3 200.0 99000.0- Removing all the rows containing 0 price since it’s
not making sense.
- Now minimum price listed $10
- 27430 unique host_id
- 42904 Total columns
- Now minimum price listed $10
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 10.0 75.0 125.0 200.4 200.0 99000.0- Minimum night
- As minimum night 1
- As average night 18
- As maximum night 1250
- As median night 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.00 2.00 7.00 18.12 30.00 1250.00- number of reviews made by host_id
- 0 to 1842
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 1.00 5.00 25.83 24.00 1842.00- Area
- Bronx has 1691 listings
- Brooklyn has 16237 listings
- Manhattan has 17658 listings
- Queens has 6916 listings
- Staten Island has 429 listings
##
## Bronx Brooklyn Manhattan Queens Staten Island
## 1690 16234 17635 6916 429- Room Types
- Entire home/apt has 24279 listings
- Hotel room has 197 listings
- Private room has 17879 listings
- Shared room has 576 listings
##
## Entire home/apt Hotel room Private room Shared room
## 24279 170 17879 576- number of total lists made in cities
area_freq <- test_com %>%
group_by(area) %>%
summarise(total_list = sum(host_list_count))%>%
mutate(percent = total_list *100 / sum(total_list))
area_freq- price range frequency
price_freq <- test_com %>%
mutate(price_range = case_when(price > 5 & price < 50 ~ "10 - 49",
price >= 50 & price < 100 ~ "50 - 99",
price >= 100 & price < 200 ~ "100 - 199",
price >= 200 & price < 300 ~ "200 - 299",
price >= 300 & price < 1000 ~ "301 - 999",
price >= 1000 ~ "above 1000")) %>%
group_by(price_range) %>%
summarise(total_list = sum(host_list_count)) %>%
mutate(percent = total_list *100 / sum(total_list)) %>%
arrange(price_range)
price_freq- total listing with price grouped by geolocation area
geo_area_freq <- test_com %>%
group_by(geo_location, area) %>%
summarise(total_list = sum(host_list_count),
min_price = min(price),
avg_price = mean(price),
max_price = max(price),
most_price = median(price))
geo_area_freq- total listing with room_type grouped by years
year_room_freq <- test_com %>%
group_by(year, room_type) %>%
summarise(total_list = sum(host_list_count),
reviews_per_year = sum(reviews_per_year))%>%
mutate(percent = 100* reviews_per_year/sum(reviews_per_year))
year_room_freq- total listings made by host_name
host_list_count_total <- test_com %>%
group_by(host_name) %>%
summarise(total_list_count = sum(host_list_count)) %>%
arrange(desc(total_list_count))
host_list_count_total- total pricing listed by host_name
host_price_total <- test_com %>%
group_by(host_name) %>%
summarise(total_price = sum(price)) %>%
arrange(desc(total_price))
host_price_total- total pricing listed by host_name by Room types
host_room_total <- test_com %>%
select(host_name, room_type, price) %>%
group_by(host_name, room_type) %>%
summarise(total_price = sum(price)) %>%
arrange(desc(total_price))## `summarise()` has grouped output by 'host_name'. You can override using the
## `.groups` argument.- room type with price by areas :
room_price_freq <- test_com %>%
select(room_type, area, price, host_list_count) %>%
group_by(room_type) %>%
summarise(min_price = min(price),
avg_price = mean(price),
most_price = median(price),
max_price = max(price),
total_list = sum(host_list_count)) %>%
mutate(list_percent = total_list * 100 / sum(total_list))
PHASE 5 : Visualization
- Creating a new data frame for Correlation matrix
- formatted as numeric
corr_df <- test_com %>%
select(list_id, host_id, price, minimum_nights,
number_of_reviews, last_review, reviews_per_month_pct,
host_list_count, reviews_per_year, availability_365, year)
corr_df$year <- as.numeric(corr_df$year)
corr_df$last_review <- as.numeric(corr_df$last_review)
str(corr_df)## 'data.frame': 42904 obs. of 11 variables:
## $ list_id : num 2595 5121 5203 5178 5136 ...
## $ host_id : num 2845 7356 7490 8967 7378 ...
## $ price : num 150 60 75 68 275 93 295 124 200 81 ...
## $ minimum_nights : num 30 30 2 2 60 3 4 3 1 30 ...
## $ number_of_reviews : num 49 50 118 575 3 350 45 223 68 189 ...
## $ last_review : num 19164 18232 17368 19407 19214 ...
## $ reviews_per_month_pct: num 0.3 0.3 0.72 3.41 0.03 2.25 0.27 1.32 0.44 1.13 ...
## $ host_list_count : int 3 2 1 1 1 1 1 3 4 1 ...
## $ reviews_per_year : int 1 0 0 52 1 48 4 17 0 5 ...
## $ availability_365 : int 314 365 0 106 181 145 1 164 310 207 ...
## $ year : num 2022 2019 2017 2023 2022 ...- Correlation matrix
- Total listing made by Host at Cities based
- Area ~ Total list Frequency
- Manhattan alone has 62.5 % of total listings
- Brooklyn has 22.1 %
- Queens has 14.6 %
- Bronx has 0.49 %
- Staten Island has 0.11 %
- Area ~ Total list Frequency
area_freq %>%
ggplot(aes(area, total_list, fill= area)) +
geom_bar(position = "dodge", stat = "identity") +
geom_text(aes(label = total_list), vjust = 0) +
guides(fill = guide_legend(title = "Area")) +
theme(legend.position = "none") +
labs(x = "Cities",
y = "Total Listings",
title = "Total Listings made at Cities :",
caption = "Data Analyst : JP") + theme_minimal() +
scale_y_continuous(labels = scales::comma) +
theme(legend.position = "none")
- Host Listing with Room Types
Room Types at NYC by Cities
- A simple visualization on total listings by price range
price_freq %>%
ggplot(aes(price_range, total_list, fill = total_list)) +
geom_bar(position = "dodge", stat = "identity") +
geom_text(aes(label = total_list), vjust = 0) +
scale_y_continuous(labels = scales::comma) +
guides(fill = guide_legend(title = "Total Lists")) +
theme(legend.position = "none") +
labs(x = "Price Range",
y = "Total Listings",
title = "Price Range for Total listings :",
subtitle = "cheap airbnb price range",
caption = "Data Analyst : JP") + theme_minimal() +
scale_y_continuous(labels = scales::comma) +
theme(legend.position = "none")## Scale for y is already present.
## Adding another scale for y, which will replace the existing scale.
- Boxplot for Room types on listed Price
- Assumed outliers as True Outliers. Since its validates ROCCC System
- we can see those outliers.
- I did limit the price range to $10,000.
ggplot(test_com, aes(x = room_type, y = price)) +
geom_boxplot(aes(fill = room_type)) + scale_y_log10(limits = c(1, 10000), labels = scales::comma) +
geom_hline(yintercept = mean(test_com$price), color = "purple", linetype = 6) +
annotate("text", x = 1,
y = median(test_com$price[test_com$room_type == "Entire home/apt"]),
label = round(median(test_com$price[test_com$room_type == "Entire home/apt"]), 2),
size = 5, color = "white") +
annotate("text", x = 2,
y = median(test_com$price[test_com$room_type == "Hotel room"]),
label = round(median(test_com$price[test_com$room_type == "Hotel room"]), 2),
size = 5, color = "red") +
annotate("text", x = 3,
y = median(test_com$price[test_com$room_type == "Private room"]),
label = round(median(test_com$price[test_com$room_type == "Private room"]), 2),
size = 5, color = "lightgreen") +
annotate("text", x = 4,
y = median(test_com$price[test_com$room_type == "Shared room"]),
label = round(median(test_com$price[test_com$room_type == "Shared room"]), 2),
size = 5, color = "green") +
theme_minimal() +
theme(legend.position = "none") +
labs(x = "Room Type",
y = "Price",
title = "Price Distribution by Room Type :",
caption = "Data Analyst : JP") ## Warning: Removed 7 rows containing non-finite values (`stat_boxplot()`).
- Room Types by Average Price Filled with Median Price Listings
- A clear view on average price vs median price listing
room_price_freq %>%
ggplot(aes(room_type, avg_price, fill = room_type)) +
geom_bar(position = "dodge", stat = "identity") +
scale_y_continuous(labels = scales::comma) +
guides(fill = guide_legend(title = "Most Price")) +
geom_text(aes(label = most_price), vjust = 0) +
theme(legend.position = "none") +
labs(x = "Room Types",
y = "Average Price",
title = "Room Types with Average Price :",
subtitle = "Median Price floating for Room Types",
caption = "Data Analyst : JP") +
theme_minimal() +
scale_y_continuous(labels = scales::comma) +
theme(legend.position = "none")## Scale for y is already present.
## Adding another scale for y, which will replace the existing scale.
- Host Listing with Average Price By Cities
Average Price Listed at NYC by Cities
- Total Listings made over Years :
- Since 2017 hosts liked the idea of Airbnb
- A clear view on listings made over period of years.
- Shared Room has almost no impact over years.
year_room_freq %>%
ggplot(aes(year, total_list, color = room_type)) +
scale_y_continuous(labels = scales::comma) +
geom_point(size = 2, alpha = 10) +
labs(x = "Years",
y = "Total Listings",
title = "Total Listings made over Years :",
subtitle = "",
caption = "Data Analyst : JP") +
theme_minimal()
- Top 5 Listings by Host Names :
- Top 1 host is Blueground created listings over 276676
host_list_count_total %>%
slice(1:5) %>%
ggplot(aes(host_name, total_list_count, fill = host_name)) +
geom_bar(position = "dodge", stat = "identity") +
geom_text(aes(label = total_list_count), vjust = 0) +
theme(legend.position = "none") +
labs(x = "Host Name",
y = "Total Listings",
title = "Top 5 Listings by Hosts",
subtitle = "",
caption = "Data Analyst : JP") +
theme_minimal() +
scale_y_continuous(labels = scales::comma) +
theme(legend.position = "none")
- Top Prices by Host Names :
- Top 1 host is RoomPicks listed total Price of
318578 $
- where Blueground is on second position for 175662 $
- Top 1 host is RoomPicks listed total Price of
318578 $
host_room_total %>%
filter( total_price >= 52089) %>%
ggplot(aes(total_price, host_name, fill = total_price)) +
geom_bar(position = "dodge", stat = "identity") +
scale_fill_steps2() +
geom_text(aes(label = total_price), vjust = 0) +
theme(legend.position = "none") +
labs(x = "Total Price in $ ",
y = "Host Name",
color = "Room Type",
title = "Top Prices by Host Names",
subtitle = "",
caption = "Data Analyst : JP") +
theme_minimal() +
theme(legend.position = "none")
- Sum of Total Price by Host Listing based on Cities :
Total Price Listed at NYC by Cities
- Reviews per month by years :
test_com %>%
ggplot(aes(reviews_per_month_pct, reviews_per_year, color = room_type)) +
geom_point(size = 2, alpha = 0.8) +
geom_smooth(method = 'lm' , se = F, color = 'purple') +
labs(x = "Reviews per Month PCT",
y = "Reviews per Year",
color = "Room type",
title = "Airbnb's Reviews per Month by Year:",
subtitle = "Linear Regression Model has 'Strog' fit",
caption = "Data Analyst : JP") +
annotate("text", x= 15, y= 900, label = "R^2 = 0.73", color = "darkgreen",
fontface = "bold", size = 5, angle = 25 ) +
theme_minimal()
PHASE 6 : Act
- The act phase would be done by the Executive team of the company. So,Passing the Documented Report to The Director and the Team.
Data-Driven Decision-Making :
Blueground $ RoomPicks are the Top 2 Hosts in NYC made 175662 $ $ 318578 $ Respc.
Blueground made maximum listings among other hosts in NYC for over 276676 listings .
Since 2017 hosts liked the idea of Airbnb, A progressive graph for Airbnb.
Hotel Rooms are the most Expensive on average price
Manhattan alone has 62.5 % of total listings**
Suggestion(s) :
Except Hotel Rooms, Rests are the cheapest, and travellers will like it
Travellers will like Bronx to stay since it can roam the city and even has lowest average price rated among others.
Staten Island has low crowd but higher average price for travellers to stay, those want less crowd can go for Staten Island
THANK YOU