What is PageRank?

PageRank is a way to measure how “important” a web page is based on how many other pages link to it—even if you’ve never studied computer science. Imagine the web as a network of roads: if many roads lead to a page, that page is like a busy intersection. PageRank gives each page a score showing how busy and well-connected it is.

Why Does It Matter?

• Search engines use it to decide which pages show up first in your search results.
• Websites track which pages are most linked to understand popularity.
• Recommender systems and social networks use similar ideas to rank content.

How PageRank Works

1. Starting Point: Picture a random web surfer who clicks links at random.
2. Following Links: With a certain chance (the damping factor), they follow a link on the current page.
3. Random Jump: The rest of the time (1 minus the damping), they jump to any page at random—this prevents the surfer from getting stuck on pages without links.

Simple Pseudocode

Steps PageRank follows:

1. Initialize each page's rank to 1/total_pages.
2. Repeat until ranks stabilize:
   a. For each page:
      i. Start with (1 - d)/total_pages.
      ii. For every page linking to it, add d * (linking_page_rank / number_of_links_on_linking_page).
3. Normalize all ranks so they sum to 1.

Where:
- d is the damping factor (e.g., 0.85).
- total_pages is the number of pages in your network.

What is the Damping Factor?

The damping factor (usually set around 0.85) tells the algorithm how often our surfer follows a link versus jumping to a random page. A high value means we trust links more, while a lower value gives more weight to random jumps.

How Links Influence the Score

Every page passes some of its importance to the pages it links to. If an important page links to you, your score goes up—just like a recommendation from a trusted friend boosts your own reputation.

Reading the Results

After running PageRank, each page gets a number between 0 and 1. Higher numbers mean the page is more central in the network. In this demo, we list pages from highest to lowest score so you can see the most “important” pages first.

Using This Demo

1. Select Custom Corpus to try built-in examples or paste your own network.
2. Select Live Website to enter a real URL (or click a sample) and explore.
3. Adjust Depth and Max Pages to control how far and wide the crawler goes.
4. Click Compute PageRank and watch both the link structure and scores update.