![[rss]](images/rss.gif)
Underestimating Digg, Or 100 Most Popular Words on Digg
One day (two days ago, as far as I remember), I asked myself a question: «What are the most popular topics on digg.com? Sex? War? $$$ per oil barrel?».
I downloaded all 571 pages of popular stories for last year with cUrl (there were more than 8000 stories), parsed them to extract titles and diggs, and processed the titles with a simple PHP script.
As you see, it is very simple. It uses Porter stemming algorithm to normalize words (so, eg. apple and apples could be recognized as the same word), but it doesn't remove prepositions, articles and other such words with no “useful meaning”. The score of each word is calculated as story's diggs multiplied by the word's relative frequency in the story title. These scores don't have any useful meaning, they are used for ordering only.
So, the top 100 digg words:
- How
- New
- Digg
- Picture
- Apple
- Free
- Top
- 10 //(remarkable, as lots of people name their articles as “10 reasons”, “10 steps”, etc...)
- Game
- Ever
- Video
- Use/using/user
- Windows
- Get
- Mac
- Amazing
- World
- Firefox
- Best
- Most
- Pic
- Year
- iPod
- Vista
- Release
- Microsoft
- About
- Time
- Image
- RIAA
- iPhone
- Internet
- Man
- Computer
- 3
- Can
- Hack
- Nintendo
- 5
- Out
- Thing
- Design
- Now
- Photoshop
- Like
- TV
- Have
- Look
- See
- Over
- Real
- Website
- Pirate
- CSS
- PS3
- Build
- Linux
- Work
- Day/Daily
- Bush
- Wait
- iTunes
- Old
- Awesome
- Show
- First
- Movie
- More
- Want
- Download
- Ad
- Online
- Say
- Site
- User
- OS
- Drive/Driver
- Cool
- PC
- Tutorial
- Take
- Flash
- Should
- Back
- Announce
- Final
- Live
- Gmail
- Own
- Bill
- 7 (another “magic:" number, like 10, 3 and 5)
- vs
- Job
- School
- Worst
- XP
- Know
- People
You understood already, what was the previous post, didn't you? I took some words from the top of the list, and pretended to be too lazy to write a good article. I was interested in how may people would click the link despite it's obvious absurdness (very few) and how many people would digg the story without following the link (none).
I thought, the social networks work much worse. Should stop underestimating them and pay more attention to them.
Top
Filed under: seo Tags:
seo,
digg.com,
popular words
