PHISHING URL DETECTION USING TRANSFORMERS

Organization Name

Microsoft Technology Licensing, LLC

Inventor(s)

Jack Wilson Stokes Iii of North Bend WA (US)

Pranav Ravindra Maneriker of Columbus OH (US)

Arunkumar Gururajan of Sammamish WA (US)

Diana Anca Carutasu of Bellevue WA (US)

Edir Vinicio Garcia Lazo of Seattle WA (US)

PHISHING URL DETECTION USING TRANSFORMERS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240297900 titled 'PHISHING URL DETECTION USING TRANSFORMERS

- Simplified Explanation:**

This technology utilizes transformers to identify phishing URLs by tokenizing useful features from the subject URL, such as text and associated data, and processing them through a transformer to determine if the URL is phishing.

- Key Features and Innovation:**
Utilizes transformers to identify phishing URLs
Tokenizes useful features from the subject URL
Includes text and associated data in the tokenization process
Generates additional training data by permuting token order and simulating attacks

- Potential Applications:**

This technology can be used in cybersecurity systems to detect and prevent phishing attacks, enhancing online security for users and organizations.

- Problems Solved:**
Identifying phishing URLs accurately
Generating additional training data for improved detection
Enhancing cybersecurity measures against phishing attacks

- Benefits:**
Improved detection of phishing URLs
Enhanced online security for users
Increased protection against cyber threats

- Commercial Applications:**

Enhancing cybersecurity systems for businesses, financial institutions, and online platforms to protect users from phishing attacks, potentially reducing financial losses and reputational damage.

- Prior Art:**

Researchers can explore prior art related to URL tokenization, transformer technology, and phishing detection algorithms to understand the evolution of this technology.

- Frequently Updated Research:**

Stay updated on advancements in transformer technology, cybersecurity algorithms, and phishing detection methods to enhance the effectiveness of this technology.

- Questions about Phishing URL Detection:**

1. How does this technology improve upon traditional methods of detecting phishing URLs? 2. What are the potential limitations of using transformers for phishing URL detection?

Original Abstract Submitted

the technology described herein can identify phishing urls using transformers. the technology tokenizes useful features from the subject url. the useful features can include the text of the url and other data associated with the url, such as certificate data for the subject url, a referrer url, an ip address, etc. the technology may build a joint byte pair encoding for the features. the token encoding may be processed through a transformer, resulting in a transformer output. the transformer output, which may be described as a token embedding, may be input to a classifier to determine whether the url is a phishing url. additional or improved url training data may be generated by permuting token order, by simulating a homoglyph attack, and by simulating a compound word attack.