Broad Listening from Text Format Data
2024-11-06
Now that the public opinion map is done, think about broad listening from text format data.
Recently I've been working on 2D visualization and clustering from opinion vectors.
The Value of Broad Listening
Is it a scatter plot? Feeling that it's not.
Most people don't know how to read a scatterplot if you just give them a scatterplot.
There is value in having an AI write the commentary.
How about just having AI provide commentary?
Someone will say, "I don't know if this explanation is correct."
I think that visualization (data, results, etc.) is just a stage direction to increase the persuasive power of AI Summary.
2025-09-09 summary felt landing on raw data
What value would an AI provide in explaining this?
To be able to understand what you didn't understand?
As an aside, the phrase "Broad listening for the masses" is misleading, and there are various forms of users, such as "those who use it for analysis" and "those who look at the results"
Nittele NEWS x 2024 House of Representatives Election x Broad Listening cases, the customer is the TV station, not the viewer, because the TV station is the one making the decision to introduce it and paying for it.
This is a complicated composition.
Customers paying TV stations are also advertisers, not viewers
The composition of the investment is such that if the topic and popularity of the new trial increases, the value of the medium will increase.
There is no such thing as broad listening where the direct mass is the customer (i.e., where the mass will pay for it).
---
This page is auto-translated from /nishio/テキスト形式データからのブロードリスニング 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.