Clip It & Ship It - smart PDF reader for better paper reading experience and knowledge mining

The goal of the Clippy project was to make a smart PDF reader for scientific papers that makes research easier and more efficient. The project was restricted to using GitHub for source code versioning and MIT-licensed libraries. All the functionalities are built upon the mandatory PDF.js library.

The result of the Clippy project should be a PDF reader that should help the researcher to parse and extract information from the paper such as tables, figures, and equations. Clippy should also build a knowledge graph out of all the references from the paper showing which papers are connected to each other. To get a quick overview of the paper Clippy should also generate a summary of the paper. All of the requirements are intended to be able to go quickly through a paper without the need to read all of it.

Initial requirements from project sponsors were:

  • resolving and presenting cross-references inside a paper
  • building a knowledge graph for the references of a paper
  • generating/mining summarizations for key components of the paper
  • (optionally) highlighting commonly used, topic-specific phrasings

Technologies used in making Clippy are:

  • Node.js -- bakcend
  • Javascript + JQuery - frontend

Contact e-mail