Final Project Solution

Automated Phishing Detection Using Machine Learning Techniques

Enter a URL and the machine learning model will classify it as Safe or Phishing with a confidence score and risk explanation.

Random Forest URL Feature Extraction Real-time Prediction

Detection TypeURL Based

BackendFlask

ModelMachine Learning

HostingcPanel Ready

Check a URL

Example: https://google.com or http://paypal-login-security.xyz

Recent Detection History

Safe

facebook.com

2026-05-14 00:17:55

68.18%

Safe

https://mail.google.com/mail/u/0/#spam

2026-05-14 00:17:47

95.0%

Safe

https://mail.google.com/mail/u/0/#spam

2026-05-14 00:17:39

95.0%

Safe

https://apd.levvn.com/api/check?url=https://google.com

2026-05-14 00:11:48

94.09%

Safe

https://apd.levvn.com/api/check?url=https://google.com

2026-05-14 00:11:36

94.09%

Safe

https://storage.googleapis.com/dafroptilanodropkingboxinggood/revokgmailinboxgoodquality.html#4zkSyw170621sdrF802ljokblvvcz99CXLKQSHCPZIDRZR86524MRHA495369y1

2026-05-14 00:10:54

74.09%

Safe

https://z3.xbx2029.com/s/15l3sVMZR9wMWT17

2026-05-14 00:10:11

92.73%

Phishing

http://verify-bank-account-update-login.ru

2026-05-14 00:09:57

99.55%

1. Feature Extraction

URL length, HTTPS, IP address, suspicious words, dots, hyphens and symbols are converted into numeric features.

2. ML Classification

A trained Random Forest model learns patterns from safe and phishing examples and predicts unknown URLs.

3. Automated Result

The app instantly shows Safe or Phishing with confidence score and explanation.