IAB Taxonomy
HOLY mackerel – it’s DEAD RIGHT.
2015 I put 32,000 of my writings through AI subject classifier and it popped out 20 short lists of words, each which represents a subject. But it didn’t name the subject, just that they’re found near each other a lot in my writings.
2020: I’m putting each of those 20 short lists of words (categories) through an IAB topic classifier made in 2019 and it says a topic with 100% certainty – and I have no doubt it’s correct.
IAB is “International Advertising Bureau”.
This is why ads have been improving since last year.
Example: If you saw list of words, what would you think the category was? What’s it talking about?
19. science history religion point truth view im part wrong culture fiction modern true world physics belief scientific reason sciences wikipedia opinion fact century en bias correct isnt theoretical psychology strong speaking considered evidence pure perspective org doesnt scientists context ideas historical beliefs suppose popular argument fine faith religions agree study agreement theories article media things generally ancient wiki debate theyre humanity perspectives religious bad research today fictional found valid western theory lot cultural pragmatic practical facts mathematics quote necessarily opinions category convince fall communities nature public studies statistics logical absolute skeptical read areas greater interesting based past stories influence rational
—
t no significance in 1-20- they’re all from list of words ‘buckets’ an AI once put my writings into about 5 years ago. This is the first time I’ve been able to find possible topics for each of the 20 short lists of words representing topics.
————–
technology and computing_artificial intelligence_26_1
is the clear winner, which you won’t see reflected above. I combined 10 of them:
6, 10, 11, 12, 13, 16, 17, 18, 19, 20
and ended up with
technology and computing_artificial intelligence_26_1
98%
religion and spirituality_atheism_21_3
2%
Contrast to when I combined the OTHER 10 together: A big mess. Very little in common.
movies_comedy movies_14_3
40%
music and audio_hard rock_15_32_4
19%
music and audio_childrens music_15_6
10%
music and audio_inspirational music_15_21
9%
music and audio_games and hobbies podcasts_15_17
9%
television_music tv_27_7
2%
video gaming_mobile games_29_3
2%
music and audio_gospel music_15_18
2%
hobbies and interests_card games_11_8_2
1%
movies_special interest movies_14_12
=====
BUT – with the removal of a SINGLE one of them #6, the percentages DRAMATICALLY shift.
religion and spirituality_atheism_21_3
94%
music and audio_religion and spirituality podcasts_15_29
4%
Where’s AI?
It’s suddenly the last 2%.
—
So this single one carries the rest 9 others into AI.
6. system systems logic knowledge work computer human programming language basic math set mathematics working data level based computers ideas understanding model general technology results ai limited project information engineering languages built fact problems thinking basis process short tool written field theory terms future machine present complete current including similar design processes area solution social common manager code approach natural task goal practical areas create skills impossible models business designed related base today operator works engineer product lead study reason education processing creative mathematical precision methods control successful resources applied accurate idea experiment allowing interest environment sources levels power external artificial
—-