Automated selection of scientifically meaningful keyword associations from literature data

Develop an automated, reproducible methodology to select scientifically meaningful associations between complex systems keywords solely from bibliographic literature data, eliminating manual curation, so that an objective keyword association network can be constructed.

Background

The paper updates the complex systems keyword diagram by constructing a weighted network of keyword associations using community feedback and bibliographic data. Due to superficial large search hits and potential misinterpretations when common terms are paired, the associations were manually reviewed and curated by the author to improve quality.

This manual curation introduces subjectivity and limits reproducibility. The paper explicitly identifies the need for a purely literature-driven approach to determine scientifically meaningful keyword associations without manual intervention, highlighting a key methodological gap for creating objective topic maps in complex systems science.