This project tries to help researchers to find references more efficiently and comprehensively. It is also a mini project for myself to get familiar with web crawler.
In the endless research life, finding nicely matching references is always a headache. You need to search and download papers many times. It is tedious and boring. And you could possible be lost in the papers, having no idea which is referencing which. How about you give some search engine a paper and it automatically returns a tree of papers showing the reference relationship? You can see how the papers are clustered and read the most influential papers first and then move to other more detailed directions.
Once this project is set up, the data obtained from the web crawler can be used to analyze and categorize the papers. "Machine reading" can be expected to help screen the most relavent papers.