Search papers, labs, and topics across Lattice.
This paper addresses the challenge of organizing and interpreting the rapidly growing biomedical literature by proposing a convex optimization-based clustering algorithm for topic modeling. The method selects exemplars from data to produce stable, fine-grained topics while guaranteeing a global optimum, thus improving reproducibility and interpretability compared to K-means, LDA, and BERTopic. Applied to 12,000 PubMed articles on aging and longevity, the method uncovers medically validated topics ranging from molecular mechanisms to lifestyle interventions.
Unlike LDA and K-means, a new convex optimization approach guarantees globally optimal and reproducible topic clusters, revealing nuanced insights from 12,000 aging research papers.
The rapid expansion of biomedical publications creates challenges for organizing knowledge and detecting emerging trends, underscoring the need for scalable and interpretable methods. Common clustering and topic modeling approaches such as K-means or LDA remain sensitive to initialization and prone to local optima, limiting reproducibility and evaluation. We propose a reformulation of a convex optimization based clustering algorithm that produces stable, fine-grained topics by selecting exemplars from the data and guaranteeing a global optimum. Applied to about 12,000 PubMed articles on aging and longevity, our method uncovers topics validated by medical experts. It yields interpretable topics spanning from molecular mechanisms to dietary supplements, physical activity, and gut microbiota. The method performs favorably, and most importantly, its reproducibility and interpretability distinguish it from common clustering approaches, including K-means, LDA, and BERTopic. This work provides a basis for developing scalable, web-accessible tools for knowledge discovery.
