import pandas as pd

import joblib

from sklearn.model_selection import train_test_split

from sklearn.ensemble import RandomForestClassifier

from sklearn.metrics import accuracy_score

import shutil

# Define file paths

data_file = '/mnt/data/urldata.csv'

features_file = '/mnt/data/features.csv'

labels_file = '/mnt/data/labels.csv'

import pandas as pd

import joblib

from sklearn.model_selection import train_test_split

from sklearn.ensemble import RandomForestClassifier

from sklearn.metrics import accuracy_score

import shutil

# Define file paths

data_file = '/mnt/data/urldata.csv'

features_file = '/mnt/data/features.csv'

labels_file = '/mnt/data/labels.csv'

import pandas as pd

import joblib
from sklearn.model_selection import train_test_split

from sklearn.ensemble import RandomForestClassifier

from sklearn.metrics import accuracy_score

import shutil

# Define file paths

data_file = '/mnt/data/urldata.csv'

features_file = '/mnt/data/features.csv'

labels_file = '/mnt/data/labels.csv'

import os

import pandas as pd

import joblib

from sklearn.model_selection import train_test_split

from sklearn.ensemble import RandomForestClassifier

from sklearn.metrics import accuracy_score

import shutil

# Define file paths

data_file = '/mnt/data/urldata.csv'

features_file = '/mnt/data/features.csv'

labels_file = '/mnt/data/labels.csv'

model_file = '/mnt/data/url_classifier.pkl'

script_file = '/mnt/data/detect_url.py'

archive_file = '/mnt/data/url_detection_project.rar'

# Load the dataset

data = pd.read_csv(data_file)
# Feature extraction function

def extract_features(url):

from urllib.parse import urlparse

import re

features = {}

# Parse URL

parsed_url = urlparse(url)

# Length Features

features['length_of_url'] = len(url)

features['length_of_hostname'] = len(parsed_url.hostname) if parsed_url.hostname else 0

features['length_of_path'] = len(parsed_url.path)

features['length_of_first_directory'] = len(parsed_url.path.split('/')[1]) if len(parsed_url.path.split('/'))

> 1 else 0

features['length_of_top_level_domain'] = len(parsed_url.hostname.split('.')[-1]) if
parsed_url.hostname else 0

# Count Features

features['count_of_-'] = url.count('-')

features['count_of_@'] = url.count('@')

features['count_of_?'] = url.count('?')

features['count_of_%'] = url.count('%')

features['count_of_.'] = url.count('.')

features['count_of_='] = url.count('=')

features['count_of_http'] = url.count('http')

features['count_of_www'] = url.count('www')

features['count_of_digits'] = sum(c.isdigit() for c in url)

 features['count_of_letters'] = sum(c.isalpha() for c in url)

features['count_of_number_of_directories'] = len(parsed_url.path.split('/')) - 1

# Binary Features

features['use_of_ip'] = 1 if re.match(r'^(http|https)://\d+\.\d+\.\d+\.\d+', url) else 0

shortening_services = (

"bit\.ly", "goo\.gl", "shorte\.st", "t\.co", "tinyurl\.com", "tr\.im", "is\.gd", "cli\.gs", "yfrog\.com",

"migre\.me", "ff\.im", "tiny\.cc", "url4\.eu", "twit\.ac", "su\.pr", "twurl\.nl", "snipurl\.com",

"short\.to",

"BudURL\.com", "ping\.fm", "post\.ly", "Just\.as", "bkite\.com", "snipr\.com", "fic\.kr", "loopt\.us",

"doiop\.com", "short\.ie", "kl\.am", "wp\.me", "rubyurl\.com", "om\.ly", "to\.ly", "bit\.do",

"lnkd\.in",

"db\.tt", "qr\.ae", "adf\.ly", "bitly\.com", "cur\.lv", "ow\.ly", "ity\.im", "q\.gs", "po\.st", "bc\.vc",

"twitthis\.com", "u\.to", "j\.mp", "buzurl\.com", "cutt\.ly", "u\.bb", "yourls\.org", "t2mio\.com",

"v\.gd",

"tr\.im", "link\.zip\.net"

pattern = re.compile(r'|'.join(shortening_services))

features['use_of_shortening_url'] = 1 if pattern.search(url) else 0

return features

# Apply feature extraction

features = data['URL'].apply(extract_features)

features_df = pd.DataFrame(features.tolist())

data = pd.concat([data, features_df], axis=1)

# Separate features and labels

X = data.drop(columns=['URL', 'Label'])

y = data['Label']
# Save the features and labels for later use

X.to_csv(features_file, index=False)

y.to_csv(labels_file, index=False)

# Split the data

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Train the model

model = RandomForestClassifier(n_estimators=100, random_state=42)

model.fit(X_train, y_train)

# Evaluate the model

y_pred = model.predict(X_test)

accuracy = accuracy_score(y_test, y_pred)

print(f'Accuracy: {accuracy}')

# Save the model

joblib.dump(model, model_file)

# Create the detect_url.py script

script_content = """

import joblib

import argparse

import pandas as pd

from urllib.parse import urlparse

# Load the model

model = joblib.load('url_classifier.pkl')
# Function to extract features (reuse the extract_features function)

def extract_features(url):

from urllib.parse import urlparse

import re

features = {}

# Parse URL

parsed_url = urlparse(url)

# Length Features

features['length_of_url'] = len(url)

features['length_of_hostname'] = len(parsed_url.hostname) if parsed_url.hostname else 0

features['length_of_path'] = len(parsed_url.path)

features['length_of_first_directory'] = len(parsed_url.path.split('/')[1]) if len(parsed_url.path.split('/'))

> 1 else 0

features['length_of_top_level_domain'] = len(parsed_url.hostname.split('.')[-1]) if
parsed_url.hostname else 0

# Count Features

features['count_of_-'] = url.count('-')

features['count_of_@'] = url.count('@')

features['count_of_?'] = url.count('?')

features['count_of_%'] = url.count('%')

features['count_of_.'] = url.count('.')

features['count_of_='] = url.count('=')

features['count_of_http'] = url.count('http')

features['count_of_www'] = url.count('www')
 features['count_of_digits'] = sum(c.isdigit() for c in url)

features['count_of_letters'] = sum(c.isalpha() for c in url)

features['count_of_number_of_directories'] = len(parsed_url.path.split('/')) - 1

# Binary Features

features['use_of_ip'] = 1 if re.match(r'^(http|https)://\\d+\\.\\d+\\.\\d+\\.\\d+', url) else 0

shortening_services = (

"bit\\.ly", "goo\\.gl", "shorte\\.st", "t\\.co", "tinyurl\\.com", "tr\\.im", "is\\.gd", "cli\\.gs",

"yfrog\\.com",

"migre\\.me", "ff\\.im", "tiny\\.cc", "url4\\.eu", "twit\\.ac", "su\\.pr", "twurl\\.nl", "snipurl\\.com",

"short\\.to",

"BudURL\\.com", "ping\\.fm", "post\\.ly", "Just\\.as", "bkite\\.com", "snipr\\.com", "fic\\.kr",

"loopt\\.us",

"doiop\\.com", "short\\.ie", "kl\\.am", "wp\\.me", "rubyurl\\.com", "om\\.ly", "to\\.ly", "bit\\.do",

"lnkd\\.in",

"db\\.tt", "qr\\.ae", "adf\\.ly", "bitly\\.com", "cur\\.lv", "ow\\.ly", "ity\\.im", "q\\.gs", "po\\.st",

"bc\\.vc",

"twitthis\\.com", "u\\.to", "j\\.mp", "buzurl\\.com", "cutt\\.ly", "u\\.bb", "yourls\\.org",

"t2mio\\.com", "v\\.gd",

"tr\\.im", "link\\.zip\\.net"

pattern = re.compile(r'|'.join(shortening_services))

features['use_of_shortening_url'] = 1 if pattern.search(url) else 0

return features

# Predict function

def predict_url(url):

features = extract_features(url)

features_df = pd.DataFrame([features])

prediction = model.predict(features_df)[0]
 return 'Malicious' if prediction == 1 else 'Benign'

# Main function for CLI

if __name__ == "__main__":

parser = argparse.ArgumentParser(description='URL Maliciousness Detection')

parser.add_argument('url', type=str, help='URL to be classified')

args = parser.parse_args()

url = args.url

result = predict_url(url)

print(f'The URL is: {result}')

"""

# Write the script to a file

with open(script_file, 'w') as f:

f.write(script_content)

# Create a RAR archive with the necessary files

shutil.make_archive('/mnt/data/url_detection_project', 'rar', '/mnt/data', '.')

# Specify the final archive file path

archive_file = '/mnt/data/url_detection_project.rar'

archive_file ​:citation[【oaicite:0】]​