Hypotheses generation using link prediction in a bipartite graph

TL;DR

This paper presents a method for automatically generating research hypotheses in physics by predicting new links between physical matter and keywords in a bipartite graph constructed from publication data, outperforming existing link prediction methods.

Contribution

It introduces a novel link prediction approach using collaborative filtering and popularity metrics to suggest new research relationships in physics literature.

Findings

Better link prediction performance than existing methods.

Effective in predicting hypotheses for specific keywords like 'antiferromagnetism' or 'superconductivity'.

Applicable to large-scale physics publication data.

Abstract

The large volume of scientific publications is likely to have hidden knowledge that can be used for suggesting new research topics. We propose an automatic method that is helpful for generating research hypotheses in the field of physics using the massive number of physics journal publications. We convert the text data of titles and abstract sections in publications to a bipartite graph, extracting words of physical matter composed of chemical elements and extracting related keywords in the paper. The proposed method predicts the formation of new links between matter and keyword nodes based on collaborative filtering and matter popularity. The formation of links represents research hypotheses, as it suggests the new possible relationships between physical matter and keywords for physical properties or phenomena. The suggested method has better performance than existing methods for link prediction in the entire bipartite graph and the subgraph that contains only a specific keyword, such as `antiferromagnetism' or `superconductivity.'