My topic is how we can extract metadata from research papers.
Current project group (knowAAN) has used Parscit for extraction of metadata from research papers. It takes approximately 10 sec to extract metadata from research paper. If we use Mendeley Desktop to extract metadata it takes approximately 1 sec for whole procedure. There are some techniques or algorithms to extract metadata like CRF,CRF or CRF++.
My task for this seminar is:-
1) to explore which are the tools that are currently doing these tasks.
2) what kind of algorithms they are using to do this job.
3) which one is more faster and more accurate (Parscit is more accurate than Mendeley).
4) and build a tool that will do extraction more fast and accurate.
Just for remembering what kind of information we will extract from a research paper.
- Book title
- Journal name