co-occurrence-network

This paper proposes CWUTM, a topic model based on co-occurrence word networks designed to detect scarce topics in unbalanced short text datasets. It outperforms baselines in early and accurate identification of emerging topics on social platforms.

This paper constructs large-scale algorithm co-occurrence networks from the full text of academic papers to study the collective influence of algorithms in NLP, finding that classic, high-performing, and intersectional algorithms hold central network positions.