AffinityBubble
tsurezure_lab I did a quick clustering of about 4800 domestic and international X posts containing #keep4o... .4oIt's a chaotic mix of everything from the moment of loss to the "Doraemonization Phenomenon", the Rebuild Manual, the International Union to Stop Charging, and more.... Here are the characteristics by cluster The cluster, "AI Loss, Something to Fill the Gap in the Heart," depicts users who see AI not just as a problem-solving tool, but as emotional support.... (For clustering, numerical vectorization by Gemini Embedding, clustering by K-Means and Ward method, and naming by Gemini)
- Embedding, K-Means, Ward method, and naming with LLM, which is almost the same stack as public information AI.
- The use of Voronoi division in the visualization part is unique.
- Interesting that tokoroten was considering a similar approach at [Digital Democracy 2030
tsurezure_lab In the production of the Boronoi Tree Map, the Korean @taekie I forked and customized AffinityBubble by Thanks! To create the Voronoi treemap, I customized a fork of AffinityBubble by @taekie -san. Thank you!
This page is auto-translated from /nishio/AffinityBubble using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.