Ccode/데이터 시각화

Netflix 데이터 간략히 EDA 해보기

맨사설 2021. 8. 5. 19:16

728x90

~~쥬피터로 아래 파일을 열어 보시면 더 예쁘게 나와요~~~

Netflix.분석.ipynb

0.14MB

실제 데이터이며 따로 분석해보실 분은 다운로드하여서 해보세요~

netflix_titles.csv

2.86MB

In [1]:

# 기본 라이브러리 불러오기
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

1. 데이터 파악하기¶

In [2]:

df = pd.read_csv('../my room/data2/netflix_titles.csv', encoding='utf-8') # 불러와서 한번 보기
df.head()

Out[2]:

	show_id	type	title	director	cast	country	date_added	release_year	rating	duration	listed_in	description
0	s1	TV Show	3%	NaN	João Miguel, Bianca Comparato, Michel Gomes, R...	Brazil	August 14, 2020	2020	TV-MA	4 Seasons	International TV Shows, TV Dramas, TV Sci-Fi &...	In a future where the elite inhabit an island ...
1	s2	Movie	7:19	Jorge Michel Grau	Demián Bichir, Héctor Bonilla, Oscar Serrano, ...	Mexico	December 23, 2016	2016	TV-MA	93 min	Dramas, International Movies	After a devastating earthquake hits Mexico Cit...
2	s3	Movie	23:59	Gilbert Chan	Tedd Chan, Stella Chung, Henley Hii, Lawrence ...	Singapore	December 20, 2018	2011	R	78 min	Horror Movies, International Movies	When an army recruit is found dead, his fellow...
3	s4	Movie	9	Shane Acker	Elijah Wood, John C. Reilly, Jennifer Connelly...	United States	November 16, 2017	2009	PG-13	80 min	Action & Adventure, Independent Movies, Sci-Fi...	In a postapocalyptic world, rag-doll robots hi...
4	s5	Movie	21	Robert Luketic	Jim Sturgess, Kevin Spacey, Kate Bosworth, Aar...	United States	January 1, 2020	2008	PG-13	123 min	Dramas	A brilliant group of students become card-coun...

In [3]:

df.info()
# director, cast, country, date_added,rating에서 null값이 있다는 것을 알 수 있다

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7787 entries, 0 to 7786
Data columns (total 12 columns):
 #   Column        Non-Null Count  Dtype 
---  ------        --------------  ----- 
 0   show_id       7787 non-null   object
 1   type          7787 non-null   object
 2   title         7787 non-null   object
 3   director      5398 non-null   object
 4   cast          7069 non-null   object
 5   country       7280 non-null   object
 6   date_added    7777 non-null   object
 7   release_year  7787 non-null   int64 
 8   rating        7780 non-null   object
 9   duration      7787 non-null   object
 10  listed_in     7787 non-null   object
 11  description   7787 non-null   object
dtypes: int64(1), object(11)
memory usage: 730.2+ KB

In [4]:

df.type.unique() # type 변수는 2개 밖에 없다는 것을 확인할 수 있다.

Out[4]:

array(['TV Show', 'Movie'], dtype=object)

In [5]:

#df.country.describe()
a=df.country.unique()
# 너무 많아 30개만 확인
for i in range(30):
    print(a[i] ,end=" ")

Brazil Mexico Singapore United States Turkey Egypt India Poland, United States Thailand Nigeria nan Norway, Iceland, United States United Kingdom Japan South Korea Italy Canada Indonesia Romania Spain Iceland South Africa, Nigeria France United States, South Africa Portugal, Spain Hong Kong, China, Singapore United States, Germany South Africa, China, United States Argentina United States, France, Serbia

In [6]:

df.duration.unique()

Out[6]:

array(['4 Seasons', '93 min', '78 min', '80 min', '123 min', '1 Season',
       '95 min', '119 min', '118 min', '143 min', '103 min', '89 min',
       '91 min', '149 min', '144 min', '124 min', '87 min', '110 min',
       '128 min', '117 min', '100 min', '2 Seasons', '84 min', '99 min',
       '90 min', '102 min', '104 min', '105 min', '56 min', '125 min',
       '81 min', '97 min', '106 min', '107 min', '109 min', '44 min',
       '75 min', '101 min', '3 Seasons', '37 min', '113 min', '114 min',
       '130 min', '94 min', '140 min', '135 min', '82 min', '70 min',
       '121 min', '92 min', '164 min', '53 min', '83 min', '116 min',
       '86 min', '120 min', '96 min', '126 min', '129 min', '77 min',
       '137 min', '148 min', '28 min', '122 min', '176 min', '85 min',
       '22 min', '68 min', '111 min', '29 min', '142 min', '168 min',
       '21 min', '59 min', '20 min', '98 min', '108 min', '76 min',
       '26 min', '156 min', '30 min', '57 min', '150 min', '133 min',
       '115 min', '154 min', '127 min', '146 min', '136 min', '88 min',
       '131 min', '24 min', '112 min', '74 min', '63 min', '38 min',
       '25 min', '174 min', '60 min', '153 min', '158 min', '151 min',
       '162 min', '54 min', '51 min', '69 min', '64 min', '147 min',
       '42 min', '79 min', '5 Seasons', '40 min', '45 min', '172 min',
       '10 min', '163 min', '9 Seasons', '55 min', '72 min', '61 min',
       '71 min', '160 min', '171 min', '48 min', '139 min', '157 min',
       '15 min', '65 min', '134 min', '161 min', '62 min', '8 Seasons',
       '186 min', '49 min', '73 min', '58 min', '165 min', '166 min',
       '138 min', '159 min', '141 min', '132 min', '52 min', '67 min',
       '34 min', '66 min', '312 min', '180 min', '47 min', '6 Seasons',
       '155 min', '14 min', '177 min', '11 min', '9 min', '46 min',
       '145 min', '11 Seasons', '7 Seasons', '13 Seasons', '8 min',
       '12 min', '12 Seasons', '10 Seasons', '43 min', '50 min', '23 min',
       '185 min', '200 min', '169 min', '27 min', '170 min', '196 min',
       '33 min', '181 min', '204 min', '32 min', '35 min', '167 min',
       '16 Seasons', '179 min', '193 min', '13 min', '214 min', '17 min',
       '173 min', '192 min', '209 min', '187 min', '41 min', '182 min',
       '224 min', '233 min', '189 min', '152 min', '19 min', '15 Seasons',
       '208 min', '237 min', '31 min', '178 min', '230 min', '194 min',
       '228 min', '195 min', '3 min', '16 min', '5 min', '18 min',
       '205 min', '190 min', '36 min', '201 min', '253 min', '203 min',
       '191 min'], dtype=object)

In [7]:

df.show_id.unique()

Out[7]:

array(['s1', 's2', 's3', ..., 's7785', 's7786', 's7787'], dtype=object)

2. 변수 선택하기¶

title, desxription 변수 같은 경우 작품마다 교유의 값을 가지므로 제외했습니다.

director, cast 변수 같은 경우 한 나라에 국한되어 있다면 공통된 값이 유의미하게 있을 수 있지만 전 세계를 대상으로 한 데이터이기에 제외했습니다.

rating 변수의 값에 대한 의미를 파악하지 못해 제외했습니다.

listed_in의 값들은 ',' 무작위 값이 많아 데이터 분석함에 복잡함을 가중할 것 같아 제외했습니다.

release_year 변수는 date_added와 비슷하며 정작 필요한 값은 date_added에서의 year값이라 생각하여 제외했습니다.

In [8]:

data = df[["type","country","date_added","duration","show_id"]] # 그렇게 선정된 5개의 변수
data.head()

Out[8]:

	type	country	date_added	duration	show_id
0	TV Show	Brazil	August 14, 2020	4 Seasons	s1
1	Movie	Mexico	December 23, 2016	93 min	s2
2	Movie	Singapore	December 20, 2018	78 min	s3
3	Movie	United States	November 16, 2017	80 min	s4
4	Movie	United States	January 1, 2020	123 min	s5

In [9]:

data.info() # country 와 release_year에 null값 존재

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7787 entries, 0 to 7786
Data columns (total 5 columns):
 #   Column      Non-Null Count  Dtype 
---  ------      --------------  ----- 
 0   type        7787 non-null   object
 1   country     7280 non-null   object
 2   date_added  7777 non-null   object
 3   duration    7787 non-null   object
 4   show_id     7787 non-null   object
dtypes: object(5)
memory usage: 304.3+ KB

In [10]:

data[data['country'].isna()]

Out[10]:

	type	country	date_added	duration	show_id
16	TV Show	NaN	March 20, 2019	1 Season	s17
38	TV Show	NaN	March 30, 2019	1 Season	s39
67	Movie	NaN	January 26, 2017	37 min	s68
97	Movie	NaN	December 31, 2019	121 min	s98
117	Movie	NaN	January 5, 2019	106 min	s118
...	...	...	...	...	...
7739	Movie	NaN	July 10, 2020	120 min	s7740
7746	TV Show	NaN	April 25, 2020	1 Season	s7747
7765	Movie	NaN	December 13, 2019	89 min	s7766
7777	TV Show	NaN	July 1, 2019	2 Seasons	s7778
7784	Movie	NaN	September 25, 2020	44 min	s7785

507 rows × 5 columns

null값 수정하기¶

In [11]:

data['country'] = data['country'].fillna(data['country'].mode()[0]) # int가 아니므로 최빈값으로 대체
data['date_added'] = data['date_added'].fillna(data['date_added'].mode()[0]) # int가 아니므로 최빈값으로 대체

<ipython-input-11-adb48114f3ea>:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  data['country'] = data['country'].fillna(data['country'].mode()[0]) # int가 아니므로 최빈값으로 대체
<ipython-input-11-adb48114f3ea>:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  data['date_added'] = data['date_added'].fillna(data['date_added'].mode()[0]) # int가 아니므로 최빈값으로 대체

In [12]:

data.info() # 수정된 것을 확인 할 수 있다.

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7787 entries, 0 to 7786
Data columns (total 5 columns):
 #   Column      Non-Null Count  Dtype 
---  ------      --------------  ----- 
 0   type        7787 non-null   object
 1   country     7787 non-null   object
 2   date_added  7787 non-null   object
 3   duration    7787 non-null   object
 4   show_id     7787 non-null   object
dtypes: object(5)
memory usage: 304.3+ KB

3. 데이터 정제 하기¶

In [13]:

data['country'] # ','로 구분하여 여러 나라가 있는 값도 있다는 것을 확인 할 수 있다.

Out[13]:

0                                                  Brazil
1                                                  Mexico
2                                               Singapore
3                                           United States
4                                           United States
                              ...                        
7782    Sweden, Czech Republic, United Kingdom, Denmar...
7783                                                India
7784                                        United States
7785                                            Australia
7786                United Kingdom, Canada, United States
Name: country, Length: 7787, dtype: object

In [14]:

data['country'] = data['country'].apply(lambda x : x.split(",")[0])
data['country']

<ipython-input-14-078caf61721c>:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  data['country'] = data['country'].apply(lambda x : x.split(",")[0])

Out[14]:

0               Brazil
1               Mexico
2            Singapore
3        United States
4        United States
             ...      
7782            Sweden
7783             India
7784     United States
7785         Australia
7786    United Kingdom
Name: country, Length: 7787, dtype: object

In [15]:

data['date_added'] # 월과 년을 사용해 볼 것이다.

Out[15]:

0          August 14, 2020
1        December 23, 2016
2        December 20, 2018
3        November 16, 2017
4          January 1, 2020
               ...        
7782      October 19, 2020
7783         March 2, 2019
7784    September 25, 2020
7785      October 31, 2020
7786         March 1, 2020
Name: date_added, Length: 7787, dtype: object

In [16]:

data['year_added'] = data['date_added'].apply(lambda x : x.split(" ")[-1])
data['month_added'] = data['date_added'].apply(lambda x : x.split(" ")[0])
data.head()

<ipython-input-16-2bf24ceab529>:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  data['year_added'] = data['date_added'].apply(lambda x : x.split(" ")[-1])
<ipython-input-16-2bf24ceab529>:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  data['month_added'] = data['date_added'].apply(lambda x : x.split(" ")[0])

Out[16]:

	type	country	date_added	duration	show_id	year_added	month_added
0	TV Show	Brazil	August 14, 2020	4 Seasons	s1	2020	August
1	Movie	Mexico	December 23, 2016	93 min	s2	2016	December
2	Movie	Singapore	December 20, 2018	78 min	s3	2018	December
3	Movie	United States	November 16, 2017	80 min	s4	2017	November
4	Movie	United States	January 1, 2020	123 min	s5	2020	January

In [17]:

data['duration'].unique()

Out[17]:

array(['4 Seasons', '93 min', '78 min', '80 min', '123 min', '1 Season',
       '95 min', '119 min', '118 min', '143 min', '103 min', '89 min',
       '91 min', '149 min', '144 min', '124 min', '87 min', '110 min',
       '128 min', '117 min', '100 min', '2 Seasons', '84 min', '99 min',
       '90 min', '102 min', '104 min', '105 min', '56 min', '125 min',
       '81 min', '97 min', '106 min', '107 min', '109 min', '44 min',
       '75 min', '101 min', '3 Seasons', '37 min', '113 min', '114 min',
       '130 min', '94 min', '140 min', '135 min', '82 min', '70 min',
       '121 min', '92 min', '164 min', '53 min', '83 min', '116 min',
       '86 min', '120 min', '96 min', '126 min', '129 min', '77 min',
       '137 min', '148 min', '28 min', '122 min', '176 min', '85 min',
       '22 min', '68 min', '111 min', '29 min', '142 min', '168 min',
       '21 min', '59 min', '20 min', '98 min', '108 min', '76 min',
       '26 min', '156 min', '30 min', '57 min', '150 min', '133 min',
       '115 min', '154 min', '127 min', '146 min', '136 min', '88 min',
       '131 min', '24 min', '112 min', '74 min', '63 min', '38 min',
       '25 min', '174 min', '60 min', '153 min', '158 min', '151 min',
       '162 min', '54 min', '51 min', '69 min', '64 min', '147 min',
       '42 min', '79 min', '5 Seasons', '40 min', '45 min', '172 min',
       '10 min', '163 min', '9 Seasons', '55 min', '72 min', '61 min',
       '71 min', '160 min', '171 min', '48 min', '139 min', '157 min',
       '15 min', '65 min', '134 min', '161 min', '62 min', '8 Seasons',
       '186 min', '49 min', '73 min', '58 min', '165 min', '166 min',
       '138 min', '159 min', '141 min', '132 min', '52 min', '67 min',
       '34 min', '66 min', '312 min', '180 min', '47 min', '6 Seasons',
       '155 min', '14 min', '177 min', '11 min', '9 min', '46 min',
       '145 min', '11 Seasons', '7 Seasons', '13 Seasons', '8 min',
       '12 min', '12 Seasons', '10 Seasons', '43 min', '50 min', '23 min',
       '185 min', '200 min', '169 min', '27 min', '170 min', '196 min',
       '33 min', '181 min', '204 min', '32 min', '35 min', '167 min',
       '16 Seasons', '179 min', '193 min', '13 min', '214 min', '17 min',
       '173 min', '192 min', '209 min', '187 min', '41 min', '182 min',
       '224 min', '233 min', '189 min', '152 min', '19 min', '15 Seasons',
       '208 min', '237 min', '31 min', '178 min', '230 min', '194 min',
       '228 min', '195 min', '3 min', '16 min', '5 min', '18 min',
       '205 min', '190 min', '36 min', '201 min', '253 min', '203 min',
       '191 min'], dtype=object)

In [18]:

data[data['duration'].str.contains('Seasons')] # Season을 포함하는 값이 많은 것을 확인할 수 있다.
# 하지만 여기서는 min과 season의 비교는 무의미 하므로 Season을 포함하는 하나의 변수를 만들겠다.

Out[18]:

	type	country	date_added	duration	show_id	year_added	month_added
0	TV Show	Brazil	August 14, 2020	4 Seasons	s1	2020	August
24	TV Show	Japan	January 23, 2020	2 Seasons	s25	2020	January
63	TV Show	United States	June 5, 2020	4 Seasons	s64	2020	June
64	TV Show	United States	August 23, 2019	3 Seasons	s65	2019	August
108	TV Show	United States	July 12, 2019	2 Seasons	s109	2019	July
...	...	...	...	...	...	...	...
7753	TV Show	Turkey	January 17, 2017	2 Seasons	s7754	2017	January
7755	TV Show	United States	January 27, 2019	5 Seasons	s7756	2019	January
7756	TV Show	Brazil	February 22, 2019	2 Seasons	s7757	2019	February
7759	TV Show	United States	September 13, 2018	3 Seasons	s7760	2018	September
7777	TV Show	United States	July 1, 2019	2 Seasons	s7778	2019	July

802 rows × 7 columns

4. 데이터 시각화 해보기¶

In [19]:

data.head()

Out[19]:

	type	country	date_added	duration	show_id	year_added	month_added
0	TV Show	Brazil	August 14, 2020	4 Seasons	s1	2020	August
1	Movie	Mexico	December 23, 2016	93 min	s2	2016	December
2	Movie	Singapore	December 20, 2018	78 min	s3	2018	December
3	Movie	United States	November 16, 2017	80 min	s4	2017	November
4	Movie	United States	January 1, 2020	123 min	s5	2020	January

In [20]:

import matplotlib.font_manager as fm
path = 'C:\\Users\\설위준\\Desktop\\my room\\NanumBarunGothic.ttf'
fontprob=fm.FontProperties(fname=path,size=18)

# Netflix type들의 분포 시각화
type_num = data.type.value_counts()
ratio=[]
for i in type_num:
    ratio.append(i)
labels = ['Movie', 'TV Show']


plt.pie(ratio,labels=labels, autopct='%.1f%%',explode=[0, 0.10], colors = ['red','lightblue'])
plt.title('Netflix 내 콘텐츠들의 분포' ,fontproperties=fontprob, fontsize = 16)
plt.show()

In [21]:

# Netflix내 5대 나라 확인하기
country_num = data.country.value_counts()
ratio=[]
for i in range(5):
    ratio.append(country_num[i])
labels=['United States','India','United Kingdom','Canada','Japen']

plt.pie(ratio,labels=labels, autopct='%.1f%%',explode=[0.10, 0.10,0.10,0.10,0.10])
plt.title('Netflix 내 5대 나라' ,fontproperties=fontprob, fontsize = 16)
plt.show()

In [22]:

# 연도별 Netflix 콘텐츠 발행 수
year_num = data.year_added.value_counts()
ratio=[]
for i in range(5):
    ratio.append(year_num[i])
labels=['2019','2020','2018','2017','2016']
plt.pie(ratio,labels=labels, autopct='%.1f%%',explode=[0.10, 0.10,0.10,0.10,0.10])
plt.title('연도별 Netflix 콘텐츠 발행 수 ' ,fontproperties=fontprob, fontsize = 16)
plt.show()

In [23]:

# 연도별 콘텐츠 발행 수
year_content = data.groupby(['year_added'])['show_id'].count().reset_index()
year_content.columns = ['Year' , 'Count']
plt.figure(figsize = (10,8))
plt.plot(year_content.Year , year_content.Count , marker = '.' , color = 'blue')
plt.title('Number of Netflix Titles released every year' , size = 16)
plt.xlabel('Year')
plt.ylabel('Count')
plt.show()

In [24]:

# 달별 Netflix 콘텐츠 발행 수 비교
month_num = data.month_added.value_counts()
ratio=[]
for i in range(5):
    ratio.append(month_num[i])
labels=['December','Octover','January','March','September']
plt.pie(ratio,labels=labels, autopct='%.1f%%',explode=[0.10, 0.10,0.10,0.10,0.10])
plt.title('달별 Netflix 콘텐츠 발행 수 ' ,fontproperties=fontprob, fontsize = 16)
plt.show()

In [25]:

from IPython.core.display import display, HTML
display(HTML("<style>.container {width:90% !important;}</style>"))

728x90

저작자표시 비영리 변경금지

'Ccode > 데이터 시각화' 카테고리의 다른 글

데이터 시각화 Tableau 로 시작하기 (0)	2021.11.14
주가 분석 (0)	2021.08.18
신용카드 사용자 연체 예측_EDA(2) (0)	2021.08.17
신용카드 사용자 연체 예측_EDA(1) (0)	2021.08.13
공공데이터를 이용한 카페 상권 분석(2021 Ver.) (0)	2021.07.31

현재글Netflix 데이터 간략히 EDA 해보기

#wannabeeeeeee the best DataScientist

Netflix 데이터 간략히 EDA 해보기

1. 데이터 파악하기¶

2. 변수 선택하기¶

null값 수정하기¶

3. 데이터 정제 하기¶

4. 데이터 시각화 해보기¶

'Ccode > 데이터 시각화' 카테고리의 다른 글

'Ccode/데이터 시각화'의 다른글

티스토리툴바

Netflix 데이터 간략히 EDA 해보기

1. 데이터 파악하기¶

2. 변수 선택하기¶

null값 수정하기¶

3. 데이터 정제 하기¶

4. 데이터 시각화 해보기¶

'Ccode > 데이터 시각화' 카테고리의 다른 글

'Ccode/데이터 시각화'의 다른글

관련글

티스토리툴바