TI18_DSBDA_4th

February 5, 2025

[5]: import pandas as pd

[6]: df=pd.read_csv("/home/dpl11/facebook.csv")

[7]: df.head()

[7]: Unnamed: 0 Page total likes Type Category Post Month Post Weekday \
0 0 139441 Photo 2 12 4
1 1 139441 Status 2 12 3
2 2 139441 Photo 3 12 3
3 3 139441 Photo 2 12 2
4 4 139441 Photo 2 12 2

Post Hour Paid Lifetime Post Total Reach \

0 3 0.0 2752
1 10 0.0 10460
2 3 0.0 2413
3 10 1.0 50128
4 3 0.0 7244

Lifetime Post Total Impressions Lifetime Engaged Users \

0 5091 178
1 19057 1457
2 4373 177
3 87991 2211
4 13594 671

Lifetime Post Consumers Lifetime Post Consumptions \

0 109 159
1 1361 1674
2 113 154
3 790 1119
4 410 580

Lifetime Post Impressions by people who have liked your Page \

0 3078
1 11710

1
 2 2812
3 61027
4 6228

Lifetime Post reach by people who like your Page \

0 1640
1 6112
2 1503
3 32048
4 3200

Lifetime People who have liked your Page and engaged with your post \
0 119
1 1108
2 132
3 1386
4 396

comment like share Total Interactions

0 4 79.0 17.0 100
1 5 130.0 29.0 164
2 0 66.0 14.0 80
3 58 1572.0 147.0 1777
4 19 325.0 49.0 393

[8]: df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 500 entries, 0 to 499
Data columns (total 20 columns):
# Column Non-
Null Count Dtype
--- ------
-------------- -----
0 Unnamed: 0 500
non-null int64
1 Page total likes 500
non-null int64
2 Type 500
non-null object
3 Category 500
non-null int64
4 Post Month 500
non-null int64
5 Post Weekday 500
non-null int64
6 Post Hour 500

2
 non-null int64
7 Paid 499
non-null float64
8 Lifetime Post Total Reach 500
non-null int64
9 Lifetime Post Total Impressions 500
non-null int64
10 Lifetime Engaged Users 500
non-null int64
11 Lifetime Post Consumers 500
non-null int64
12 Lifetime Post Consumptions 500
non-null int64
13 Lifetime Post Impressions by people who have liked your Page 500
non-null int64
14 Lifetime Post reach by people who like your Page 500
non-null int64
15 Lifetime People who have liked your Page and engaged with your post 500
non-null int64
16 comment 500
non-null int64
17 like 499
non-null float64
18 share 496
non-null float64
19 Total Interactions 500
non-null int64
dtypes: float64(3), int64(16), object(1)
memory usage: 78.2+ KB

[9]: df.shape

[9]: (500, 20)

[10]: df.dtypes

[10]: Unnamed: 0 int64

Page total likes int64
Type object
Category int64
Post Month int64
Post Weekday int64
Post Hour int64
Paid float64
Lifetime Post Total Reach int64
Lifetime Post Total Impressions int64
Lifetime Engaged Users int64

3
 Lifetime Post Consumers int64
Lifetime Post Consumptions int64
Lifetime Post Impressions by people who have liked your Page int64
Lifetime Post reach by people who like your Page int64
Lifetime People who have liked your Page and engaged with your post int64
comment int64
like float64
share float64
Total Interactions int64
dtype: object

[11]: df.isnull().sum()

[11]: Unnamed: 0 0
Page total likes 0
Type 0
Category 0
Post Month 0
Post Weekday 0
Post Hour 0
Paid 1
Lifetime Post Total Reach 0
Lifetime Post Total Impressions 0
Lifetime Engaged Users 0
Lifetime Post Consumers 0
Lifetime Post Consumptions 0
Lifetime Post Impressions by people who have liked your Page 0
Lifetime Post reach by people who like your Page 0
Lifetime People who have liked your Page and engaged with your post 0
comment 0
like 1
share 4
Total Interactions 0
dtype: int64

[12]: #Creating the subset of data

df_subset1 = df[['like','share']]

[13]: df_subset1

[13]: like share

0 79.0 17.0
1 130.0 29.0
2 66.0 14.0
3 1572.0 147.0
4 325.0 49.0
.. … …

4
 495 53.0 26.0
496 53.0 22.0
497 93.0 18.0
498 91.0 38.0
499 91.0 28.0

[500 rows x 2 columns]

[14]: df_subset2 = df[['comment','Type']]

[15]: df_subset2

[15]: comment Type

0 4 Photo
1 5 Status
2 0 Photo
3 58 Photo
4 19 Photo
.. … …
495 5 Photo
496 0 Photo
497 4 Photo
498 7 Photo
499 0 Photo

[500 rows x 2 columns]

[16]: #Merging the DataFrames

merged_data = pd.merge(df_subset2, df_subset1, left_on='comment',␣
↪right_on='like')

merged_data

[16]: comment Type like share

0 4 Photo 4.0 2.0
1 4 Photo 4.0 1.0
2 4 Photo 4.0 0.0
3 4 Photo 4.0 1.0
4 4 Status 4.0 2.0
… … … … …
1462 56 Photo 56.0 17.0
1463 56 Photo 56.0 8.0
1464 56 Photo 56.0 12.0
1465 56 Photo 56.0 9.0
1466 56 Photo 56.0 25.0

[1467 rows x 4 columns]

5
[17]: # Sort by 'Likes' in descending order
sorted_df = df.sort_values(by='like', ascending=False)
print("\nSorted Data by Likes (Descending):")
print(sorted_df)

Sorted Data by Likes (Descending):

Unnamed: 0 Page total likes Type Category Post Month Post Weekday \
244 244 130791 Photo 2 7 3
379 379 111620 Photo 3 4 1
349 349 117764 Photo 3 5 5
168 168 135428 Photo 1 9 3
3 3 139441 Photo 2 12 2
.. … … … … … …
21 21 138414 Photo 1 12 7
100 100 137020 Photo 1 10 4
441 441 98195 Photo 1 3 5
417 417 104070 Photo 1 3 3
111 111 136736 Photo 1 10 6

Post Hour Paid Lifetime Post Total Reach \

244 5 1.0 180480
379 14 1.0 105632
349 13 0.0 81856
168 10 0.0 41984
3 10 1.0 50128
.. … … …
21 10 0.0 1384
100 9 1.0 1357
441 4 1.0 1845
417 10 0.0 1874
111 8 0.0 1261

Lifetime Post Total Impressions Lifetime Engaged Users \

244 319133 8072
379 147918 3984
349 124753 3000
168 68290 3370
3 87991 2211
.. … …
21 2467 15
100 2453 37
441 2670 9
417 2474 25
111 2158 37

Lifetime Post Consumers Lifetime Post Consumptions \

6
244 4010 6242
379 2254 3391
349 1637 2718
168 2420 4074
3 790 1119
.. … …
21 15 20
100 37 55
441 9 9
417 25 31
111 37 49

Lifetime Post Impressions by people who have liked your Page \

244 108752
379 48575
349 52477
168 34802
3 61027
.. …
21 2196
100 2154
441 1614
417 1483
111 1911

Lifetime Post reach by people who like your Page \

244 51456
379 27328
349 27392
168 20928
3 32048
.. …
21 1172
100 1120
441 1008
417 1062
111 1077

Lifetime People who have liked your Page and engaged with your post \
244 3316
379 1936
349 1756
168 2126
3 1386
.. …
21 15
100 32
441 9

7
 417 15
111 33

comment like share Total Interactions

244 372 5172.0 790.0 6334
379 51 1998.0 128.0 2177
349 45 1639.0 122.0 1806
168 144 1622.0 208.0 1974
3 58 1572.0 147.0 1777
.. … … … …
21 0 0.0 0.0 0
100 0 0.0 0.0 0
441 0 0.0 0.0 0
417 0 0.0 0.0 0
111 0 NaN NaN 0

[500 rows x 20 columns]

[18]: # Transpose the DataFrame

merged_data.transpose()

[18]: 0 1 2 3 4 5 6 7 8 \
comment 4 4 4 4 4 4 4 4 4
Type Photo Photo Photo Photo Status Status Status Status Status
like 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0
share 2.0 1.0 0.0 1.0 2.0 1.0 0.0 1.0 2.0

9 … 1457 1458 1459 1460 1461 1462 1463 1464 \

comment 4 … 51 51 51 146 146 56 56 56
Type Status … Photo Photo Photo Photo Photo Photo Photo Photo
like 4.0 … 51.0 51.0 51.0 146.0 146.0 56.0 56.0 56.0
share 1.0 … 11.0 6.0 6.0 9.0 15.0 17.0 8.0 12.0

1465 1466
comment 56 56
Type Photo Photo
like 56.0 56.0
share 9.0 25.0

[4 rows x 1467 columns]

[19]: merged_data.T

[19]: 0 1 2 3 4 5 6 7 8 \
comment 4 4 4 4 4 4 4 4 4
Type Photo Photo Photo Photo Status Status Status Status Status
like 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0

8
 share 2.0 1.0 0.0 1.0 2.0 1.0 0.0 1.0 2.0

9 … 1457 1458 1459 1460 1461 1462 1463 1464 \

comment 4 … 51 51 51 146 146 56 56 56
Type Status … Photo Photo Photo Photo Photo Photo Photo Photo
like 4.0 … 51.0 51.0 51.0 146.0 146.0 56.0 56.0 56.0
share 1.0 … 11.0 6.0 6.0 9.0 15.0 17.0 8.0 12.0

1465 1466
comment 56 56
Type Photo Photo
like 56.0 56.0
share 9.0 25.0

[4 rows x 1467 columns]

[20]: #Shape
df.shape

[20]: (500, 20)

[21]: # Reshape
df_temp = pd.DataFrame({'foo': ['one', 'one', 'one', 'two', 'two', 'two'],

'bar': ['A', 'B', 'C', 'A', 'B', 'C'],

'baz': [1, 2, 3, 4, 5, 6],
'zoo': ['x', 'y', 'z', 'q', 'w', 't']})

[22]: df_temp

[22]: foo bar baz zoo

0 one A 1 x
1 one B 2 y
2 one C 3 z
3 two A 4 q
4 two B 5 w
5 two C 6 t

[23]: import matplotlib.pyplot as plt

import seaborn as sns
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import accuracy_score,confusion_matrix
from sklearn.linear_model import LogisticRegression
import seaborn as sns
import matplotlib.pyplot as plt
# import pandas library
import numpy as np

9
[25]: def remove_outliers(column):
Q1 = column.quantile(0.25)
Q3 = column.quantile(0.75)
IQR = Q3 - Q1
threshold = 1.5 * IQR
outlier_mask = (column < Q1 - threshold) | (column > Q3 + threshold)
return column[~outlier_mask]

[27]: df = df.drop_duplicates()

[29]: col_name = ['Category','Post Month','Post Weekday','Post Hour','Paid']

for col in col_name:
df[col] = remove_outliers(df[col])

[30]: plt.figure(figsize=(10, 6)) # Adjust the figure size if needed

for col in col_name:
sns.boxplot(data=df[col])
plt.title(col)
plt.show()

10
11
12
[ ]:

13