⁵⁰ views
ubuntu2004

Kernel: Python 3 (system-wide)

Website Analysis using python

In [2]:

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

import warnings
warnings.filterwarnings('ignore')

plt.style.use('fivethirtyeight')
%matplotlib inline

In [3]:

dataset = pd.read_csv('3dlookmewebsite.csv', encoding='latin')

In [36]:

dataset.head()

Out[36]:

Quick Statistic Analysis

In [4]:

dataset.describe()

Out[4]:

Dealing with data types

In [14]:

dataset.dtypes

Out[14]:

User-ID                                                  float64
Gender                                                    object
Age                                                      float64
Month wise progress with DSM at 3dlook.me website         object
Number of Customers                                       object
Customer Retention Rate                                   object
Time spent (minutes)                                     float64
Engagement Rate                                           object
Increase in Users                                        float64
CTR                                                       object
Bounce Rate                                               object
Net Promoter Score                                       float64
Average Order Value                                       object
Sales                                                     object
Visit Number                                             float64
Items Purchased                                          float64
Overall customer satisfaction                            float64
Online Ambiance Review                                    object
Customer Loyalty Change                                   object
Users Referred                                           float64
Sensory Method                                            object
Time of Day                                               object
Marketing Channels                                        object
dtype: object

In [15]:

dataset.isnull().sum()

Out[15]:

User-ID                                                  25
Gender                                                   25
Age                                                      25
Month wise progress with DSM at 3dlook.me website         2
Number of Customers                                       2
Customer Retention Rate                                   2
Time spent (minutes)                                      2
Engagement Rate                                           2
Increase in Users                                         2
CTR                                                       2
Bounce Rate                                               2
Net Promoter Score                                        2
Average Order Value                                       2
Sales                                                     2
Visit Number                                              2
Items Purchased                                           2
Overall customer satisfaction                             2
Online Ambiance Review                                   17
Customer Loyalty Change                                   2
Users Referred                                            9
Sensory Method                                           22
Time of Day                                               7
Marketing Channels                                        2
dtype: int64

In [17]:

dataset[dataset['Online Ambiance Review'].isnull()].head()

Out[17]:

Checking out columns separately

In [18]:

dataset['Online Ambiance Review'].value_counts().head()

Out[18]:

Good product                                                            37
Excellent product                                                       17
Nice product                                                            11
Awesome product                                                          8
Facebook post about the entertaining projection mapping in the store     3
Name: Online Ambiance Review, dtype: int64

In [28]:

item_counts = dataset['Online Ambiance Review'].value_counts().sort_values(ascending=False).iloc[0:15]
sns.barplot(x=item_counts.index, y=item_counts.values, palette=sns.cubehelix_palette(15))
plt.ylabel("Counts")
plt.title("Which items were bought more often?");
plt.xticks(rotation=90);

Out[28]:

In [30]:

dataset['Online Ambiance Review'].value_counts().tail()

Out[30]:

"The website's use of videos and animations is so fun and engaging"                           1
"The website's music and colors are so calming and enjoyable, it feels like a nice escape"    1
4                                                                                             1
Positive comment on a YouTube video about the informative product videos in the stor          1
Positive comment on a YouTube video about the informative product videos in the store         1
Name: Online Ambiance Review, dtype: int64

Checking out columns separately

In [19]:

dataset.head()

Out[19]:

In [22]:

item_counts = dataset['Online Ambiance Review'].value_counts().sort_values(ascending=False).iloc[0:15]
plt.figure(figsize=(18,6))
sns.barplot(x=item_counts.index, y=item_counts.values, palette=sns.cubehelix_palette(15))
plt.ylabel("Counts")
plt.title("Which items were bought more often?");
plt.xticks(rotation=90);

Out[22]:

In [23]:

dataset['Customer Loyalty Change'].value_counts().tail()

Out[23]:

Increased by 0.22    2
Increased by 0.98    2
Increased by 0.65    2
Increased by 0.87    2
Increased by 0.88    2
Name: Customer Loyalty Change, dtype: int64

In [24]:

dataset['Overall customer satisfaction'].value_counts().tail()

Out[24]:

9     1
8     1
7     1
6     1
7    1
Name: Overall customer satisfaction, dtype: int64

In [25]:

dataset['Customer Retention Rate'].value_counts().tail()

Out[25]:

20%    1
90%    1
60%    1
30%    1
80%    1
Name: Customer Retention Rate, dtype: int64

In [29]:

dataset['Net Promoter Score'].value_counts().tail()

Out[29]:

3     1
6     1
9     1
2     1
7    1
Name: Net Promoter Score, dtype: int64

In [38]:

dataset['Number of Customers'].value_counts().tail()

Out[38]:

60%     1
00%     1
40%     1
90%     1
00%    1
Name: Number of Customers, dtype: int64

In [39]:

dataset['Customer Retention Rate'].value_counts().tail()

Out[39]:

20%    1
90%    1
60%    1
30%    1
80%    1
Name: Customer Retention Rate, dtype: int64

In [40]:

dataset['Time spent (minutes) '].value_counts().tail()

Out[40]:

0      1
0     1
0     1
0     1
0    1
Name: Time spent (minutes) , dtype: int64

In [41]:

dataset['Engagement Rate'].value_counts().tail()

Out[41]:

80%    1
20%    1
40%    1
20%    1
80%    1
Name: Engagement Rate, dtype: int64

In [42]:

dataset['Increase in Users'].value_counts().tail()

Out[42]:

0    1
0    1
0    1
0    1
0    1
Name: Increase in Users, dtype: int64

In [45]:

# replace NaN values with empty string
dataset['Online Ambiance Review'] = dataset['Online Ambiance Review'].fillna('')

# filter out rows where the review is in full upper case
lcase_counts = dataset[~dataset['Online Ambiance Review'].str.isupper()]['Online Ambiance Review'].value_counts().sort_values(ascending=False).iloc[0:15]

# plot the results
plt.figure(figsize=(18,6))
sns.barplot(x=lcase_counts.index, y=lcase_counts.values, palette=sns.color_palette("hls", 15))
plt.ylabel("Counts")
plt.title("Not full upper case items")
plt.xticks(rotation=90);

Out[45]:

In [5]:

dataset['Customer Loyalty Change'].value_counts().head()

Out[5]:

Increased by 0.91    7
Increased by 0.62    6
Increased by 0.57    5
Increased by 0.76    5
Increased by 0.89    5
Name: Customer Loyalty Change, dtype: int64

In [8]:

customer_loy = dataset['Customer Loyalty Change'].value_counts().sort_values(ascending=False).iloc[0:15]
plt.figure(figsize=(18,6))
sns.barplot(x=customer_loy.index, y=customer_loy.values, palette=sns.color_palette("GnBu_d"))
plt.ylabel("Counts")
plt.title("Customer loyalty change rate ")
plt.xticks(rotation=90)

Out[8]:

(array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14]),
 [Text(0, 0, 'Increased by 0.91'),
  Text(1, 0, 'Increased by 0.62'),
  Text(2, 0, 'Increased by 0.57'),
  Text(3, 0, 'Increased by 0.76'),
  Text(4, 0, 'Increased by 0.89'),
  Text(5, 0, 'Increased by 0.52'),
  Text(6, 0, 'Increased by 0.34'),
  Text(7, 0, 'Increased by 0.78'),
  Text(8, 0, 'Increased by 0.81'),
  Text(9, 0, 'Increased by 0.43'),
  Text(10, 0, 'Increased by 0.67'),
  Text(11, 0, 'Increased by 0.24'),
  Text(12, 0, 'Decreased by 0.18'),
  Text(13, 0, 'Increased by 0.39'),
  Text(14, 0, 'Increased by 0.28')])

In [0]:

Website Analysis using python

Quick Statistic Analysis

Dealing with data types

Checking out columns separately

Checking out columns separately

Product

Resources

Company