aswinssoman / extracting-core-claims

An attempt to extract single core claims from scientific articles

Extracting-Core-Claims

An attempt to extract single core claims from scientific articles

In this project, an attempt is made to extract single core claims from scientific articles.

The code works as follows:

First, the core sentence is extracted from a scientific article. This is done in Extract_Sentence_abstract.py Currently, for performance reasons, we have chosen to only look at the abstract for extracting the core sentence.

When the core sentence is extracted, AIDA comes into place to check sentences. AIDA is a concept brought to life by Kuhn et al. [1] and offers a way to structure scientific claims. It is proposed as a tool for researchers to easily access and communicate research hypotheses, claims and opinions. The rules of an AIDA sentence are used in this approach to determine whether we are dealing with a single claim or not.

Therefore, the second step is to check whether the core sentence complies with the rules of an AIDA sentence. If it does, a single claim is extracted from the article. If it does not, an attempt is made to rewrite the sentence in order to make it comply with these rules. This is all done in AIDA_Check_and_Rewrite.py

The results and all other required files for reproduction of the project are also provided, feel free to use them for further improvements of the algorithm!

Cheers,

Tom

[1] Kuhn, Tobias, et al. "Broadening the scope of nanopublications." The Semantic Web: Semantics and Big Data. Springer Berlin Heidelberg, 2013. 487-501.