lelandmcinnes.bsky.social
A Mathematician dabbling in Data Science, especially unsupervised learning and data exploration. UMAP, HDBSCAN, PyNNDescent, DataMapPlot. (He/Him)
👤 Joined on 25 March 2024
Followers
Following
Total Posts
@lelandmcinnes.bsky.socialCould also be on e.g. linearized wasserstein embeddings between texts based on word embeddings from SVD of the PMI matrix (I think that's what @lelandmcinnes.bsky.social talks about here, IIRC, good talk nevertheless) www.youtube.com/watch?v=Z6s7...
Show PostLeland McInnes built UMAP to help humans (not AI) make sense of messy piles of images, documents, and video. His full thinking on data, taste, and exploration is on The Test Set. Listen now: https://posit.co/thetestset/episode/smuggled-gold-and-untrustworthy-intuition-with-leland-mcinnes
Show PostDiscover how to build a text clustering pipeline using large language model embeddings and HDBSCAN. Automatically uncover topics in unlabeled text data with advanced clustering techniques. 📚🔍 #NLP #Clustering #MachineLearning
Show PostWe ran 3 different unsupervised ML methods on the exact same symptom matrix: A) autoencoder + HDBSCAN B) ensemble UMAP + k-means consensus C) latent class analysis (LCA) Then we asked how much they actually agreed. #LongCovid 3/
Show PostClustering Unstructured Text with LLM Embeddings and HDBSCAN
Show Post