In [160]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
In [198]:
data1 = pd.read_csv('C:/My Files/Python Practice/internet_session.csv', parse_dates=['start_time'])
In [ ]:
In [199]:
data1.shape
Out[199]:
(4712, 9)
In [181]:
data1.isna().sum()
Out[181]:
name                     0
start_time               0
usage_time               0
IP                       0
MAC                      0
upload                   0
download                 0
total_transfer           0
seession_break_reason    9
dtype: int64
In [200]:
data1 = data1.dropna().copy()
data1.isna().sum()
Out[200]:
name                     0
start_time               0
usage_time               0
IP                       0
MAC                      0
upload                   0
download                 0
total_transfer           0
seession_break_reason    0
dtype: int64
In [201]:
data1.duplicated().sum()
Out[201]:
0
In [202]:
data1 = data1.dropna().copy()
data1.isna().sum()
Out[202]:
name                     0
start_time               0
usage_time               0
IP                       0
MAC                      0
upload                   0
download                 0
total_transfer           0
seession_break_reason    0
dtype: int64
In [203]:
data1['usage_time'] = data1['usage_time'].str.replace('00:', '', 1)
data1['usage_time'] = pd.to_datetime(data1['usage_time'])

data1['upload'] = data1['upload'].str.extract('(\d+)', expand=False)
data1_usage.upload = data1.upload.astype(float)

data1['download'] = data1['download'].str.extract('(\d+)', expand=False)
data1.download = data1.download.astype(float)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_34384\3553692667.py in <module>
      3 
      4 data1['upload'] = data1['upload'].str.extract('(\d+)', expand=False)
----> 5 data1_usage.upload = data1.upload.astype(float)
      6 
      7 data1['download'] = data1['download'].str.extract('(\d+)', expand=False)

NameError: name 'data1_usage' is not defined
In [204]:
# Create an empty list to store the device names
device = []

# Check if the 'MAC' column exists in the dataframe
if 'MAC' in data1.columns:
    # Assign the value of the first MAC address in the 'MAC' column to the variable 'MAC'
    MAC = data1['MAC'][0]
    device_number = 1
    for i in data1['MAC']:
        # Check if the current MAC address is the same as the previous one
        if i == MAC:
            # Append the device name to the list
            device.append(basename + str(device_number))
        else:
            # Increment the device number
            device_number += 1
            # Append the new device name to the list
            device.append(basename + str(device_number))
            MAC = i
    # Create a new column in the dataframe called 'device'
    data1['device'] = device
else:
    print("No 'MAC' column found in the dataframe.")
In [205]:
data1.dtypes
Out[205]:
name                             object
start_time               datetime64[ns]
usage_time               datetime64[ns]
IP                               object
MAC                              object
upload                           object
download                         object
total_transfer                  float64
seession_break_reason            object
device                           object
dtype: object
In [206]:
data1['hour'] = pd.to_datetime(data1['start_time']).dt.hour
frequent_activity_time_of_day = data1['hour'].value_counts().sort_index()

What is the most frequent internet activity time of the day ?¶

In [190]:
plt.figure(figsize=(18, 9))
sns.lineplot(data=frequent_activity_time_of_day)
plt.xticks(np.linspace(start=0, stop=24, num=25))
plt.show()
plt.clf()

<Figure size 640x480 with 0 Axes>
In [192]:
print('The maximum usage time is:')
print(data1.usage_time.max())

The maximum usage time is:
2023-01-28 22:00:07

How often the ip changes ?¶

In [195]:
base_ip = '48:E7:DA:58:22:E9'
ip_count = 0
for i in range(1, data1.shape[0]):
    if data1.iloc[i]['IP'] != base_ip:
        ip_count +=1
        base_ip = data1.iloc[i]['IP']

print('The IP Adress changed ' + str(ip_count) + ' times') 

The IP Adress changed 2303 times

How often the device changed¶

In [207]:
base_device = 'device1'
device_count = 0
for i in range(1, data1.shape[0]):
    if data1.iloc[i]['device'] != base_device:
        device_count +=1
        base_device = data1.iloc[i]['device']

print('The device changed ' + str(device_count) + ' times')

The device changed 1223 times

What is the average usage per hour , per day and per month¶

In [209]:
data1.reset_index(inplace=True)

data1['day'] = data1['start_time'].dt.day
data1['month'] = data1['start_time'].dt.month

hourly_average = data1.groupby('hour').total_transfer.mean()
print('The Average usage per hour is:\n ' +  str(round(hourly_average, 2)))

The Average usage per hour is:
 hour
0     464530.44
1     530880.86
2     431576.11
3     345303.34
4     359809.44
5     275960.91
6     468959.59
7     292886.83
8     366681.92
9     377480.64
10    393259.12
11    309492.45
12    310137.98
13    335270.58
14    472403.71
15    517005.11
16    403919.40
17    525423.69
18    666590.76
19    389841.79
20    355862.80
21    474038.34
22    449600.50
23    407785.08
Name: total_transfer, dtype: float64
In [212]:
daily_average = data1.groupby('day').total_transfer.mean()
print('The Average usage per day is:\n ' +  str(round(daily_average, 2)))

The Average usage per day is:
 day
1     396705.04
2     494496.48
3     445865.63
4     676332.03
5     652195.66
6     396261.75
7     402259.89
8     301859.57
9     393521.97
10    350665.02
11    729857.65
12    346695.95
13    501906.70
14    352701.10
15    521520.51
16    426719.39
17    475795.71
18    337490.93
19    301941.32
20    365130.12
21    462211.69
22    486595.37
23    383153.93
24    320598.94
25    443689.47
26    463432.02
27    324318.12
28    494576.34
29    363645.61
30    361418.88
31    369118.01
Name: total_transfer, dtype: float64
In [213]:
monthly_average = data1.groupby('month').total_transfer.mean()
print('The Average usage per month is:\n ' +  str(round(monthly_average, 2)))

The Average usage per month is:
 month
5     311177.16
6     338418.08
7     418583.99
8     479042.44
9     482955.52
10    549467.63
11    399804.11
Name: total_transfer, dtype: float64