Skip to content

This problem statement given in internship project provided by iNeuron. The mail goal is to predict whether the domains are real or malicious.

Notifications You must be signed in to change notification settings

ujjwalkar0/Phishing-Domain-Detection

Repository files navigation

Phishing-Domain-Detection

Abstract:

An attacker who is impersonating someone or a company in order to obtain sensitive information like login credentials or account information is called phishing. Phishing is most popular amoung cybercriminals because it is easier to persuade someone to click a malicious link than to compromise a computer. Using machine learning, analysing features of website I have build a Random Forest classification model which gives 96% accuracy to predict if domain is phishing or not ?

Problem Statement:

Phishing is a type of fraud in which an attacker impersonates a reputable company or person in order to get sensitive information such as login credentials or account information via email or other communication channels. Phishing is popular among attackers because it is easier to persuade someone to click a malicious link that appears to be authentic than it is to break through a computer's protection measures.

The mail goal is to predict whether the domains are real or malicious.

Solution:

A labeled public dataset from the Mendaly Data comprised of features of 88,647 website is used for prediction. Clasification is done with Random Forest Classification Algorithm gives 96% accuracy.

Dataset:

A labeled public dataset from the Mendaly Data comprised of features of 88,647 website is used for prediction

URL: https://data.mendeley.com/datasets/72ptz43s9v/1

About

This problem statement given in internship project provided by iNeuron. The mail goal is to predict whether the domains are real or malicious.

Resources

Stars

Watchers

Forks