The First-Level Categories of Yahoo Directories
|
Arts & Humanities Recreation & Sports Education Science Health |
News & Media Computers & Internet Regional Government Society & Culture |
Business & Economy Reference Entertainment Social Science |
Three URLs with Keywords, Descriptions, and Hyperlinks
| URL | Information | |
|---|---|---|
| URL1 | Keyword | baseball, MLB, bat, college baseball, home run, news |
| Description | Baseball.com – catch your Major League Baseball info | |
| Hyperlink | baseball (22), sports (12), news (3), computer (2) | |
| URL2 | Keyword | basketball, NBA, block, defense, foul out, NBA news |
| Description | Basketball! NBA news, basketball sites, NBA scores | |
| Hyperlink | basketball (17), football (6), sports (6), news (2) | |
| URL3 | Keyword | football, NFL, National Football League, block, defense |
| Description | Football! Chat room for football fans, football sites | |
| Hyperlink | football (33), sports (9) | |
An Example
Classify web pages by using a height-three modified decision tree which splits the root, depth-one nodes, and depth-two nodes based on keywords, descriptions, and hyperlinks, respectively. A classification starts with a web page at the root of the decision tree and traces paths downward to leaves, which give the categories of the page.
|
Johnny collected lots of money from trick or treating and he went to the candy store to buy some chocolate. “You should give that money to charity”, said the shopkeeper. “No, I’ll buy the chocolate. YOU give the money to charity!” |