Researchers have evaluated unsupervised methods for recommending scholarly collaborations based on publication text. The study compared TF-IDF, topic-based models (LDA, BERTopic), and embedding-based retrieval using SciBERT with Faiss. Results indicated that topic and embedding-based approaches maintained stable performance even when publication overlap was reduced, suggesting they capture broader similarities beyond direct lexical matches. The paper also explored explainability through intrinsic topic-based and post-hoc retrieval-based methods, offering complementary insights. AI
RANK_REASON The cluster contains an academic paper detailing research findings on scholarly collaboration recommendations. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
