Accommodations in Sri Lanka

Here I used  Kaggle website (https://www.kaggle.com) data file of Sri Lanka accommodations for this analyzing report. I used python jupyter Notebook for this report analyze.

• for the data set - Accommodations in SL.csv file
• for complete codes - Jupyter Notebook code

Sri Lanka tourism industry is rapidly trending. So I thought analyze these type of accommodations type data. I don't know that time period which is the taken these data, but at the end this was a good help for my analytics skills.

Importing libraries & Importing the file

In [2]:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

In [4]:

df = pd.read_csv("Information for Accommodation.csv")
df

Out[4]:

	Type	Name	Address	Rooms	Grade	District	AGA Division	PS/MC/UC	Logitiute	Latitude
0	Boutique Hotels	THE THEVA RESIDENCY	11/B5/10-1 06TH LANE, HANTANA,KANDY	10	NaN	Kandy	Kandy Divisional Secretariat	Kandy	80.635411	7.276036
1	Boutique Hotels	HIGHLAND VILLA	351, ABIMANGAMA ROAD, KUMBALGAMA, WELIGAMA	10	NaN	Matara	Weligama Divisional Secretariat	Weligama Pradeshiya Sabha	80.409972	5.960334
2	Boutique Hotels	ULAGALLA WALAWWA RESORT	THIRAPPANE, ANURADHAPURA	21	NaN	Anuradhapura	Anuradhapura East	Anuradhapura	80.545063	8.205927
3	Boutique Hotels	GALLE FORT HOTEL	NO.28, CHURCH STREET, FORT , GALLE	14	NaN	Galle	Galle Divisional Secretariat	Galle	80.217563	6.026649
4	Boutique Hotels	THE ELEPHANT CORRIDOR	POTHANA,KIBISSA, SIGIRIYA	21	NaN	Matale	NaN	Matale	80.710743	7.943525
...	...	...	...	...	...	...	...	...	...	...
2125	Tourist Hotels	HOTEL NIPPON	123, KUMARANRATNAM ROAD, COLOMBO 02	49	NaN	Colombo	Colombo	Colombo Divisional Secretariat	79.850099	6.923932
2126	Tourist Hotels	THIDAS ARANA HOTEL	23RD,MILE POST,BANDIWEWA,POLONNARUWA	11	NaN	Polonnaruwa	Thamankaduwa Pradeshiya Sabha	Thamankaduwa Divisional Secretariat	80.986056	7.951938
2127	Tourist Hotels	THE GRAND MOUNTAIN HOTEL	73/1,ROHANA DISSANAYAKE MAWATHA,MC ROAD,MATALE.	86	NaN	Matale	Matale	Matale Divisional Secretariat	NaN	NaN
2128	Tourist Hotels	ARIE LAGOON - NEGOMBO	587,PAMUNUGAMA ROAD,THALAHENA, NEGOMBO	55	NaN	Gampaha	Negombo	Negombo Divisional Secretariat	79.823391	7.168803
2129	Tourist Hotels	THE VILLAS	352/4,RATHNAYAKA ROAD,THALPITIYA SOUTH,WADDUWA,,	32	NaN	Kalutara	Panadura Pradeshiya Sabha	Panadura Divisional Secretariat	79.923348	6.670724

2130 rows × 10 columns

Spreading

First I use the bar plot to visualize the data how spread the accommodations according to the each district.

In [161]:

x = df.groupby(['District']).count().index

y = df.groupby(['District']).count()['Type']

y = [int(df.strip()) if type(df) == str else df for df in y]

plt.figure(figsize = (28,8), dpi = 500)

plt.title('Number of Accommodation')

bar = plt.bar(x,y, color ='red')

bar[4].set_hatch('///')

plt.savefig('')

plt.show()

And after we can see here high frequency belongs to the Colombo district, and it shown above within dash lines. In after respectively Galle, Kandy, Gampaha, Kaluthara can be mentioned.

Accommodation Grades

Basically we can identify 11 varieties of grades. Which are ('A', 'B', 'C', 'DELUXE', 'FIVE', 'FOUR', 'ONE', 'STANDARD', 'SUPERIOR','THREE', 'TWO')

In [13]:

grades = ['A', 'B', 'C', 'DELUXE', 'FIVE', 'FOUR', 'ONE', 'STANDARD', 'SUPERIOR','THREE', 'TWO']

a = [df.loc[df['Grade'] == "A"].count()[0],
    df.loc[df['Grade'] == grades[1]].count()[1],
    df.loc[df['Grade'] == grades[2]].count()[1],
    df.loc[df['Grade'] == grades[3]].count()[1],
    df.loc[df['Grade'] == grades[4]].count()[1],
    df.loc[df['Grade'] == grades[5]].count()[1],
    df.loc[df['Grade'] == grades[6]].count()[1],
    df.loc[df['Grade'] == grades[7]].count()[1],
    df.loc[df['Grade'] == grades[8]].count()[1],
    df.loc[df['Grade'] == grades[9]].count()[1],
    df.loc[df['Grade'] == grades[10]].count()[1]]

explode = (0,0,.3,0,.6,.8,.2,0,0,.7,.3)

plt.style.use('default')

plt.figure(figsize =(233,11))

plt.pie(a, labels = grades, autopct = '%.3f%%', pctdistance = 0.8, explode = explode )

plt.legend()

plt.show()

When set up to a pie chart we can see basically most of accommodations have reported into two grades (A-25.095% & B-21.72%) as top and three grades are in below the top class as getting nearly equal percentages (DELUXE-14.861%, STANDARD-14.643%, SUPERIOR-14.154%)

Here we can see the lowest percentage as FOUR grade (1.034%) which is lack of distribution of accommodation in Sri Lanka.

Colombo Accommodations

In [ ]:

#A Grade Accommodations

A_accom_in_col = df.loc[(df['District'] == 'Colombo') & (df['Grade'] == 'A')]
print(A_accom_in_col.count()[0])

In [86]:

#FOUR Grade Accommodation

Four_accom_in_col = df.loc[(df['District'] == 'Colombo') & (df['Grade'] == 'FOUR')]
print(Four_accom_in_col['Address'])

432    NO. 200, UNION PLACE, COLOMBO 02
436    NO. 100, HOTEL ROAD, MT. LAVINIA
Name: Address, dtype: object

Not for each we can see how spreading Colombo Accommodations. So we can get in one code and we have 281 accommodations in around to Colombo with varies type of grades.

In [9]:

accom_in_col = df.loc[(df['District'] == 'Colombo') & (df['Grade'])]
accom_in_col

Out[9]:

	Type	Name	Address	Rooms	Grade	District	AGA Division	PS/MC/UC	Logitiute	Latitude
70	Bangalows	THE PIER	NO 113/8B ,DUWAWATTA, ST PETER'S ROAD, KORALAW...	4	DELUXE	Colombo	Moratuwa	Moratuwa Divisional Secretariat	79.888801	6.773335
76	Bangalows	THE RANMIN BUNGALOW	NO.9, TUDUWA, DAMPE, PILIYANDALA.	3	SUPERIOR	Colombo	Colombo	Colombo Divisional Secretariat	79.918606	6.766099
97	Bangalows	ARANGA-LA	NO.10,HOSPITAL ROAD,HOMAGAMA	3	DELUXE	Colombo	Homagama Pradeshiya Sabha	Homagama Divisional Secretariat	NaN	NaN
106	Bangalows	THAMBARAVILA	"THAMBARAVILA",NO.44/5,5TH LANE,NAWALA ROAD,RA...	2	SUPERIOR	Colombo	Kotte Muncipal Council	Kotte Divisional Secretariat	NaN	NaN
109	Bangalows	THE OCEAN FRONT	14, SELLAMUTTU AVENUE, COLOMBO 3	6	DELUXE	Colombo	Colombo	Colombo Divisional Secretariat	79.849100	6.908351
...	...	...	...	...	...	...	...	...	...	...
1892	Rented Apartments	CITY BEDS	NO. 20 3/1, REGENT BUILDING,,PARSONS ROAD,COLO...	3	SUPERIOR	Colombo	Colombo	Colombo Divisional Secretariat	NaN	NaN
1893	Rented Apartments	HAWK HOLIDAYS	20 2/3, REGENT BUILDING,REGENT FLATS, PARSONS ...	3	STANDARD	Colombo	Colombo	Colombo Divisional Secretariat	NaN	NaN
1894	Rented Apartments	AV 15	NO. 18, ALEXANDRA ROAD, COLOMBO 06	24	DELUXE	Colombo	Colombo	Colombo Divisional Secretariat	NaN	NaN
1899	Rented Homes	SAVENRA GARDEN RESORT	13A, RATHABALAHENA ESTATE, PADUKKA ROAD, GODAG...	5	SUPERIOR	Colombo	Homagama Pradeshiya Sabha	Homagama Divisional Secretariat	80.036651	6.850166
1900	Rented Homes	COZY BEDS	NO 29/ 2A, CHAPEL ROAD, NUGEGODA.	4	SUPERIOR	Colombo	Kotte Muncipal Council	Nugegoda Divisional Secretariat	79.892414	6.872081

281 rows × 10 columns

In [10]:

#Grouping & Counting

accom_in_col.groupby(['Grade']).count()

Out[10]:

	Type	Name	Address	Rooms	District	AGA Division	PS/MC/UC	Logitiute	Latitude
Grade
A	58	58	58	58	58	58	58	39	40
B	49	49	49	49	49	49	49	16	16
C	2	2	2	2	2	2	2	0	0
DELUXE	53	53	53	53	53	53	53	30	30
FIVE	5	5	5	5	5	5	5	5	5
FOUR	2	2	2	2	2	2	2	2	2
ONE	10	10	10	10	10	10	10	10	10
STANDARD	26	26	26	26	26	26	26	10	10
SUPERIOR	62	62	62	62	62	62	62	24	24
THREE	9	9	9	9	9	9	9	9	9
TWO	5	5	5	5	5	5	5	5	5

By grouping Colombo accommodation grades we can see here 62 SUPERIOR grades accommodations. That is the most spreading grade in Colombo district. Only 2 FOUR grade accommodation ad that are located, NO. 200, UNION PLACE, COLOMBO 02' (HILTON COLOMBO RESIDENCE), 'NO. 100, HOTEL ROAD, MT. LAVINIA' (MOUNT LAVINIA HOTEL).

We can show the in simple bar chart as follows.

In [11]:

Accommodations_in_col = accom_in_col.groupby(['Grade']).count()['Type']

Accommodations_in_col = [int(Accommodations_in_col.strip(''))
if type(Accommodations_in_col) == str 
else Accommodations_in_col for Accommodations_in_col in Accommodations_in_col]

In [14]:

plt.figure(figsize = (12,6))

bar = plt.bar(grades, Accommodations_in_col, color = 'orange')

bar[0].set_hatch('///')
bar[1].set_hatch('ooo')
bar[3].set_hatch('--')
bar[8].set_hatch('...')


plt.xlabel('Accommodation Grade')
plt.ylabel('Number of Accommodations')
plt.title('GRADES TYPES OF COLOMBO ACCOMMODATIONS')



plt.show()

According to this bar plot we can identify generally A, B, DELUXE and SUPERIOR grades accommodations have mostly spreaded in Colombo district.

Number of Rooms Distribution in Sri Lanka Accommodations

Basically in a accommodation we know it should have any number of rooms. In rapid inspection I could identify some of accommodations have only one room and nearly five rooms as well as Sri Lanka accommodations have more than 500 rooms. So we can analyze how they are varies with ranges of room numbers.

In [153]:

rooms = [df.loc[(df['Rooms'] < 5)].count()[1],
         df.loc[(df['Rooms'] >=5) & (df['Rooms'] < 10)].count()[1],
         df.loc[(df['Rooms'] >=10) & (df['Rooms'] < 20)].count()[1],
         df.loc[(df['Rooms'] >=20) & (df['Rooms'] < 30)].count()[1],  
        df.loc[(df['Rooms'] >=30) & (df['Rooms'] < 50)].count()[1],
        df.loc[(df['Rooms'] >=50) & (df['Rooms'] < 100)].count()[1],
        df.loc[(df['Rooms'] >=100) & (df['Rooms'] < 200)].count()[1],
        df.loc[(df['Rooms'] >=200) & (df['Rooms'] < 300)].count()[1],
        df.loc[(df['Rooms'] >=300) & (df['Rooms'] < 400)].count()[1],
        df.loc[(df['Rooms'] >=400) & (df['Rooms'] < 500)].count()[1],
        df.loc[(df['Rooms'] >=500)].count()[1]]
        
Num_of_rooms = ['< 5','5-10','10-20','20-30','30-50','50-100','100-200','200-300','300-400','400-500','> 500']

explode = (0,0,0,0,0,0,0,0,0.1,0.3,0.6)

plt.style.use('default')

plt.figure(figsize =(233,11))

plt.pie(rooms, labels = Num_of_rooms, autopct = '%.1f%%', pctdistance = 0.4, explode = explode )

plt.title("NUMBER OF ROOMS DISTRIBUTION IN SRI LANKA ACCOMODATIONS")

plt.legend()

plt.show()

In here we can see Sri Lanka accommodations have mostly five to ten rooms. That percentage is 31.4%. Not only that range less than five room range also have nearly that percentage which is 30.2%. when we get that values as totally we have 61.6% percentage value as less than 10 rooms in Sri Lanka accommodations. And also more than 400 rooms we can see just its only 0.1%. And when clearly inspecting that range we can see they are 'GALADARI HOTEL','RIU HOTEL','CINNAMON GRAND HOTEL' & 'SHANGRI- LA HOTEL COLOMBO'. And all are classified as 'Classified Hotels( 1-5 Star)'.

Classifications of Accommodations

Mainly Sri Lanka accommodations are divided into 11 category in this data file. So I grouped and counted as follows.

In [9]:

df.groupby(df['Type']).count()['Name']

Out[9]:

Type
Bangalows                       348
Boutique Hotels                  28
Boutique Villas                  36
Classified Hotels( 1-5 Star)    141
Guest Houses                    894
Heritage Bungalows                4
Heritage Homes                    2
Home Stay Units                 378
Rented Apartments                64
Rented Homes                      6
Tourist Hotels                  229
Name: Name, dtype: int64

Basically here we can identify most of accommodations in Sri Lanka is the Guest House. That is 894 out of 2130 accommodations. And also Bungalows & Home Stay Units are equally distributed with 300+ accommodations. And the especial case of this we have 141 (1-5-stars) hotels.

As well as we have 6 Heritage accommodations. Two of them are Heritage Homes, others are Heritage Bungalows.(we can see follow details of them)

In [25]:

df.loc[(df['Type'] == 'Heritage Bungalows') | (df['Type'] == 'Heritage Homes')]

Out[25]:

	Type	Name	Address	Rooms	Grade	District	AGA Division	PS/MC/UC	Logitiute	Latitude
1447	Heritage Bungalows	ELPITIYA WALAUWA	ELPITIYA ESTATE,WELIGALLA.	5	DELUXE	Kandy	Gampola	Gampaha Divisional Secretariat	80.602462	7.190628
1448	Heritage Bungalows	BROCKENHURST	98, WATER FIELD DRIVE, NUWARA ELIYA	4	DELUXE	Nuwara Eliya	Nuwara Eliya	Nuwara Eliya Divisional Secretariat	80.764033	6.981032
1449	Heritage Bungalows	BETHANY 101	101, COLOMBO ROAD, PUTTALAM.	6	DELUXE	Puttalam	Puttalam Pradeshiya Sabha	Puttalam Divisional Secretariat	79.830405	8.027126
1450	Heritage Bungalows	HERONDALE BUNGALOW	HERONDALE ESTATE, A2 ESTATE - BLOCK 02, NAGAST...	4	DELUXE	Kandy	Kandy	Kandy Divisional Secretariat	NaN	NaN
1451	Heritage Homes	ATAPATTU WALAWWA	35/5, ATAPATTU WALAWWA, LOWER DICKSON ROAD, GALLE	7	DELUXE	Galle	Galle	Galle Divisional Secretariat	80.217628	6.038721
1452	Heritage Homes	WILWATTE WALAWWA	WILWATTE WALAWWA, WILWATTE, MIRIGAMA	1	SUPERIOR	Gampaha	mirigama	Mirigama Divisional Secretariat	80.126954	7.242735

Simply we can set up a pie chart for this data

In [24]:

Names = df.groupby(df['Type']).count()['Name'].index
Types = df.groupby(df['Type']).count()['Name']
Types = [df.strip('')
         if type(Types) == str
         else df 
         for df in Types]

plt.style.use('fivethirtyeight')
plt.figure(figsize = (15,15))
explo = (0,0.09,0,0,0,0.1,0.35,0,0,0.6,0)
plt.pie(Types, labels = Names, autopct = '%.2f%%', pctdistance = .8, explode= explo)
plt.title('CLASSIFICATION OF SRI LANKA ACCOMMODATIONS')
plt.savefig("CLASSIFICATION OF SRI LANKA ACCOMMODATIONS", dpi= 300)
plt.show()