To run the project:

1. The project is made in Eclipse.
2. Build and run the project.
3. Enter input "N" to get top N websites as output.

Any questions regarding building and running the project can be sent to vaibhav.hemant@gmail.com 

LOGIC:

1. Pull the list of 500 candidate sites from the given link (http://topsites.eaiti.com/).
2. Create a Map with a candidate link as a key and an Integer value.
3. The Integer value of the key corresponds to number of incoming links to the website.
4. Start crawling the candidate websites. 
5. The constant value CRAWLING_LIMIT can be adjusted to increase the rank confidence(with increase in crawling time).
6. Make a GET request and gather all the nested HTML links in the <a href> tags.
7. If a link on the parent URL is a link to one of the candidate keys, increment the corresponding "incoming link value".
8. Sort the <CandidateURL, incomingLinks> Map based on the incomingLinks value.
9. Return the top N websites found.

Note 1:When I started measuring for "QUALITY" of links along with the "QUANTITY" of links, the results were not acceptable. For eg, "BLOGGER.COM" was consistently ranked as #1..
Thereafter, I measured only the number of links.

Output Example:

With CRAWLER_LIMIT set to 1, following were the results for N=10

Website Rank #1 amazon.com
Website Rank #2 google.com
Website Rank #3 t.co
Website Rank #4 facebook.com
Website Rank #5 twitter.com
Website Rank #6 youtube.com
Website Rank #7 instagram.com
Website Rank #8 baidu.com
Website Rank #9 stackexchange.com
Website Rank #10 oracle.com

With CRAWLER_LIMIT set to 50, following were the results for N=10

Website Rank #1 google.com
Website Rank #2 amazon.com
Website Rank #3 t.co
Website Rank #4 facebook.com
Website Rank #5 twitter.com
Website Rank #6 youtube.com
Website Rank #7 instagram.com
Website Rank #8 baidu.com
Website Rank #9 stackexchange.com
Website Rank #10 oracle.com