AdGraph: A Graph-Based Approach to Ad and Tracker Blocking(Background Knowledge)

안녕하세요 :D

 

I will review a paper titled "ADGRAPH: A Graph-Based Approach to Ad and Tracker Blocking."

 

When I first read through the paper from beginning to end, I found the content difficult to grasp. I realized I lacked some background knowledge on the topic, so I decided to do some additional research to better understand the context before revisiting the paper.

---

Here is the background information I gatherd:

 

Overview of Online Adverstising and Tracking

Today, most targeted advertising relies on online tracking techniques.

To understand how ads are able to identify and follow users, I first looked into how the advertising ecosystem works.

 

Key Players in the Ad Ecosystem

• Publisher: The owner of the webpage the user visits
• Advertiser: The party wanting to display advertisements
• Ad Network: A mediator that connects Publishers and Advertisers

Example Process:

1. Advertisers use an Ad Network to insert ad-related JavaScript or images into the Publisher’s webpage
2. When a user visits that page
3. The browser loads the ad or tracking content provided by the advertiser

 

Online Tracking

How do ads know what products I’m interested in and recommend them to me?

This is achieved by tracking the user’s browsing history to understand their preferences.
Tracking typically involves identifying which websites the user visits, allowing trackers to build a behavioral profile.

 

Cookies and Cookie Syncing

• Cookie: A small piece of data stored by a website on the user’s browser to remember information
  (e.g., keeping a user logged in or remembering items in a shopping cart)
• Third-Party Cookie: A cookie set by a domain different from the one the user is visiting
  (e.g., while visiting example.com, a cookie from ads.com is set)
• Cookie Syncing:
  A technique where multiple tracking companies share their cookie data to identify the same user across different websites

Example of Cookie Syncing

1. tracker.com places a cookie user123 via website1.com
2. advertiser.com sets a different cookie userABC via website2.com for the same user
   

   

3. When the user visits website3.com, which includes an image from tracker.com, a redirect occurs to advertiser.com
4. As a result, advertiser.com receives the browser request along with its own cookie userABC and can match it with user123 from tracker.com
   → This allows them to synchronize cookie identifiers and recognize the user as the same person

 

Privacy Concerns

Through cookie syncing and online tracking, a user’s entire web activity can be quietly monitored.
This often happens without the user’s awareness, raising serious privacy concerns.

---

With this background knowledge, I was better able to understand how the ADGRAPH approach proposed in the paper detects and blocks trackers through a graph-based methodology.