MBTI Text Analyzer

[ClevelandINTP]

Found this site that interprets the cognitive functions you're using in your writing style: https://www.uclassify.com/browse/g4mes543/myers-briggs-type-indicator-text-analyzer?input=Text

I fed this thing about 5-6 texts varying in length of 1-5K words that I have wrote. Apparently, my finished projects are VERY INTJ

 

[BritNi]

Fun!

This thing is bias against INFJs though.
Apparently it's just like the rest of the world which doesn't understand them.

Lol.

 

[ClevelandINTP]

Fun!

This thing is bias against INFJs though.
Apparently it's just like the rest of the world which doesn't understand them.

Lol.

How so?

I think it depends on the context of the work you're feeding it. For example, I used some of my finished projects which are very direct, forceful, analytical, insightful and future-oriented. Most likely were heavily influenced by corporate preferences. Perhaps, that is associated with the language of the INTJ

Yet, when I feed it my brief insights from a twitter-esc feed, I am typed ENTP. Apparently, ENTP is the language of quick, insightful, witty, punchy, scattered and analytical

I think your observation is interesting

 

[Lady Jolanda]

I have fed the classifier several different texts over various contexts, and it burbs out everything from INTJ to INTP to ENTP to ENFP to INFP to ESFJ to ENFJ. ¯\_(ツ)_/¯

 

[ClevelandINTP]

We do use all of the cognitive functions, though. Seems like it classifies your writing style as more 'NT'

Frankly, all of my "polished" work is classified as INTJ. These are long texts, too, so more data, central limit theorem, etc.

Doesn't mean I'm INTJ, but because it's more complete and thought through, I'm sure my writing is super-concise, action-oriented and focused on future direction

 

[BritNi]

It's not a valid tool or anything.
I think it's more for fun.

 

[BritNi]

How so?

I think it depends on the context of the work you're feeding it. For example, I used some of my finished projects which are very direct, forceful, analytical, insightful and future-oriented. Most likely were heavily influenced by corporate preferences. Perhaps, that is associated with the language of the INTJ

Yet, when I feed it my brief insights from a twitter-esc feed, I am typed ENTP. Apparently, ENTP is the language of quick, insightful, witty, punchy, scattered and analytical

I think your observation is interesting

Lol. I'm sure if you write about conspiracy theories and aliens and fortune telling, it would type you as INFJ.

Lol.

 

[ClevelandINTP]

First paragraph of catcher in the rye picks up on heavy Fi influence

ISFP
25%
ENFP
15%
ENFJ
12%
INFP
11%
INTJ
7%
ESFJ
6%
ISTP
6%
ESTP
4%
ISFJ
4%
ISTJ

 

[ClevelandINTP]

It's not a valid tool or anything.
I think it's more for fun.

Valid and accurate are two different things

Also, the whole subjectivity of things

 

[ClevelandINTP]

I really wouldn't underestimate the application of deep-learning when it comes to text transcription

 

[BritNi]

It's not an underestimation rather than some skepticism.

Before I would ever take something like this as more than *fun*, I would really like to know the credentials of those who are creating the algorithms, ya know?

Valid and accurate are two different things, yes. Thanks for pointing that out. Perhaps there is some accuracy within this tool, at times. But I highly doubt it's valid.

Just my opinion and all.

 

[ClevelandINTP]

It's not an underestimation rather than some skepticism.

Before I would ever take something like this as more than *fun*, I would really like to know the credentials of those who are creating the algorithms, ya know?

Valid and accurate are two different things, yes. Thanks for pointing that out. Perhaps there is some accuracy within this tool, at times. But I highly doubt it's valid.

Just my opinion and all.

Fair enough. I think our definitions of valid and accurate are swapped. I think accuracy requires validity whereas validity doesn't always mean accuracy

 

[acd]

Fun!

This thing is bias against INFJs though.
Apparently it's just like the rest of the world which doesn't understand them.

Lol.

I fed it your post and this was the result:

INFJ
25%
ISFP
13%
INFP
12%
ESTJ
11%
ENTJ
8%
ENFP
7%
INTP
6%
INTJ
4%
ENTP
4%
ESFJ
3%
ISFJ
2%
ESFP
2%
ISTP
2%
ESTP
1%
ENFJ
1%
ISTJ
0%

lol!

But I agree with you that this is all a bit silly.

 

[BritNi]

Fair enough. I think our definitions of valid and accurate are swapped. I think accuracy requires validity whereas validity doesn't always mean accuracy

Reliable is the word I'm thinking of. Not valid.

Silly me.

 

[ClevelandINTP]

It's not an underestimation rather than some skepticism.

Before I would ever take something like this as more than *fun*, I would really like to know the credentials of those who are creating the algorithms, ya know?

Valid and accurate are two different things, yes. Thanks for pointing that out. Perhaps there is some accuracy within this tool, at times. But I highly doubt it's valid.

Just my opinion and all.

Here are the culprits: https://www.uclassify.com/about

 

[ClevelandINTP]

I fed it your post and this was the result:


lol!

But I agree with you that this is all a bit silly.

I mean, it's scooping up a lot of "IF" and more "Fi" followed by dominant "Te" users

Which is basically an INFP. Perhaps their algorithm could use some slight tweaking but it seems like it's getting the essence...

A better question to ask is: is this classifier better at picking up type than just random selection and or is it better than a human could do

 

[BritNi]

I fed it your post and this was the result:


lol!

But I agree with you that this is all a bit silly.

Lol. That's funny.

I wonder if it just fed INFJ because it included "INFJ". Hmm... Lemme check.


Lol.

I removed "against INFJs" and it fed me this:


Which I find humor in because it kind of supports my hypothesis.

 

[Asa]

An interesting and fun experiment. Thanks for posting this @ClevelandINTP.

I tried different chapters of my novels and got INTJ most often.
I tried different clips of dialogue and earned the correct type for the character each time. Interesting. I wonder if this is a good tool for checking authenticity when trying to write "typed" dialogues.
Note: I typed the characters after writing, not before.

 

[ClevelandINTP]

An interesting and fun experiment. Thanks for posting this @ClevelandINTP.

I tried different chapters of my novels and got INTJ most often.
I tried different clips of dialogue and earned the correct type for the character each time. Interesting. I wonder if this is a good tool for checking authenticity when trying to write "typed" dialogues.
Note: I typed the characters after writing, not before.

Neat! Yeah, I don't think it's perfect, but it's has some accuracy within it. It's certainly providing me a nice mid-day mind-fuck

Imagine we're in a simulation and everything we've ever read has been created to give the illusion of what a certain type might have wrote and or thought

 

[ClevelandINTP]

I mean, it's scooping up a lot of "IF" and more "Fi" followed by dominant "Te" users

Which is basically an INFP. Perhaps their algorithm could use some slight tweaking but it seems like it's getting the essence...

A better question to ask is: is this classifier better at picking up type than just random selection and or is it better than a human could do

Oh, nevermind. I see what you did