Let the LLM explain the PCA axes.

- 2024-09-08-Technology to Support Democracy on [Polis-like visualization of the 2022 House of Councillors election - Around 1h22 - Talk about PCA’s first principal component axis coinciding with right. - > Bluemo I’m curious about the vertical axis. - Good idea. - What is the second principal component?

Chat about this comment:.

• 66e709acaff09e0000254d7e
  • In a simple conflict, the contribution of the first principal component should be larger

  • In other words, whether the distribution of arguments is plurality/multidimensional/velvety or one-dimensional/step should be quantifiable by principal component analysis.

It would be interesting to create a contribution rate for each axis, its direction vector, and a “natural language explanation” for the high-dimensional opinion vector distribution.

---

This page is auto-translated from /nishio/PCAの軸をLLMに説明させる 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.