# Profiling Users by Modeling Web Transactions

## TL;DR

This paper presents a method to profile individual users based on their web transaction behaviors using feature extraction and one-class classification, effectively distinguishing among 25 users over six months of web traffic data.

## Contribution

It introduces a novel user profiling technique leveraging web transaction features and one-class classifiers, demonstrating its effectiveness and efficiency in real-world network data.

## Key findings

- Successfully differentiated 25 users with high accuracy
- Effective in real-world network traffic over six months
- Fast and scalable profiling method

## Abstract

Users of electronic devices, e.g., laptop, smartphone, etc. have characteristic behaviors while surfing the Web. Profiling this behavior can help identify the person using a given device. In this paper, we introduce a technique to profile users based on their web transactions. We compute several features extracted from a sequence of web transactions and use them with one-class classification techniques to profile a user. We assess the efficacy and speed of our method at differentiating 25 users on a dataset representing 6 months of web traffic monitoring from a small company network.