Semantic search + open-status classifier on ResearchMath-14k
Complete NLP pipeline for research mathematics: TF-IDF extraction, sentence embeddings, UMAP visualization, K-Means clustering, semantic search, and binary classification of problem open status.
• Field-specific keyword extraction via TF-IDF
• Sentence embedding generation and visualization
• Unsupervised clustering to map problem landscape
• Semantic search to find related problems
• Open-status prediction classifier
• Near-duplicate detection by similarity