FredZhang7 commited on
Commit
15f79b5
1 Parent(s): 6988e01

Create plots.py

Browse files
Files changed (1) hide show
  1. plots.py +52 -0
plots.py ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import pandas as pd
2
+ import matplotlib.pyplot as plt
3
+ import seaborn as sns
4
+ import numpy as np
5
+
6
+ # Combine "combined_df.csv" and "combined_val_df.csv" into one dataframe
7
+ df = pd.concat([pd.read_csv('phishing_features_train.csv'), pd.read_csv('phishing_features_val.csv')], ignore_index=True)
8
+
9
+ # Define the columns to plot
10
+ columns_to_plot = ['redirects', 'not_indexed_by_google', 'issuer', 'certificate_age', 'email_submission', 'request_url_percentage', 'url_anchor_percentage', 'meta_percentage', 'script_percentage', 'link_percentage', 'mouseover_changes', 'right_click_disabled', 'popup_window_has_text_field', 'use_iframe', 'has_suspicious_port', 'external_favicons', 'TTL', 'ip_address_count', 'TXT_record', 'check_sfh', 'count_domain_occurrences', 'domain_registeration_length', 'abnormal_url', 'age_of_domain', 'page_rank_decimal']
11
+
12
+ # Create a list to store the file names of the saved plots
13
+ file_names = []
14
+
15
+ # Loop through the columns and create the scatterplot or barplot
16
+ for column in columns_to_plot:
17
+ if df[column].dtype == 'int64' or df[column].dtype == 'float64':
18
+ fig, ax = plt.subplots()
19
+ sns.regplot(x=column, y='is_malicious', data=df, ax=ax)
20
+ corr_coef = df[[column, 'is_malicious']].corr().iloc[0,1]
21
+ ax.set_title(f'{column} vs is_malicious\nCorrelation Coefficient: {corr_coef:.2f}')
22
+ file_name = f'{column}_scatterplot.png'
23
+ plt.savefig(file_name)
24
+ file_names.append(file_name)
25
+ elif df[column].dtype == 'object':
26
+ fig, ax = plt.subplots()
27
+ if (df[column] == "None").sum() > 0:
28
+ sns.countplot(x=column, hue='is_malicious', data=df[df[column] == "None"], ax=ax)
29
+ ax.set_title(f'{column} (null) vs is_malicious')
30
+ file_name = f'{column}_null_barplot.png'
31
+ plt.savefig(file_name)
32
+ file_names.append(file_name)
33
+ sns.countplot(x=column, hue='is_malicious', data=df, ax=ax)
34
+ ax.set_title(f'{column} (all) vs is_malicious')
35
+ file_name = f'{column}_all_barplot.png'
36
+ plt.savefig(file_name)
37
+ file_names.append(file_name)
38
+
39
+ # Create a figure with subplots to combine the saved plots
40
+ num_plots = len(file_names)
41
+ num_rows = int(np.ceil(num_plots/2))
42
+ fig, axs = plt.subplots(num_rows, 2, figsize=(20, 5*num_rows))
43
+ for i, file_name in enumerate(file_names):
44
+ row = i // 2
45
+ col = i % 2
46
+ img = plt.imread(file_name)
47
+ axs[row, col].imshow(img)
48
+ axs[row, col].axis('off')
49
+ if num_plots % 2 == 1:
50
+ axs[num_rows-1, 1].axis('off')
51
+ plt.tight_layout()
52
+ plt.savefig('correlation_coefficient.png')