Timeline for K-means clustering of word embedding gives strange results
Current License: CC BY-SA 3.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 11, 2018 at 19:51 | comment | added | Russell Richie | Paragram-SL999 are a large set of word embeddings that get nearly human performance on a word-word similarity benchmark. You might get better results with these: cs.cmu.edu/~jwieting | |
| Apr 27, 2018 at 6:20 | answer | added | Has QUIT--Anony-Mousse | timeline score: 17 | |
| Apr 27, 2018 at 1:15 | comment | added | Emre | Antonyms are still similar by distribution, or context. If your goal is to separate them, try a different model, such as LWET: Revisit Word Embeddings with Semantic Lexicons for Modeling Lexical Contrast. Welcome to the site! | |
| Apr 27, 2018 at 0:38 | history | asked | Thusitha | CC BY-SA 3.0 |