Using AI to solve the Beale ciphers

[chalkoutline]

Within another thread, Advanceforgold suggested using AI to solve the ciphers which makes a lot of sense.so started a new thread for this topic.

Advanceforgold also suggested using python.

 

[chalkoutline]

Within another thread, Advanceforgold suggested using AI to solve the ciphers which makes a lot of sense.so started a new thread for this topic.

Here are my thoughts...
I think the starting point would be to use the uncorrected DOI string; "ihaledeposotedinthecopnttol..."

And then have AI separate and correct the message.
"I have deposited in the county of.."

Once AI can do that, then the next step would be to bring in a "text" document to remove punctuation, headers, the spaces, etc. Documents before 1820 puts the first word of the next at the bottom. That too needs to be removed.

Then do the same for different formats: pdfs, jpgs, etc. There are probably converter to do this that could be leveraged: pdf to text, jpg to text, etc.

 

[Juskevicius]

Here are my thoughts...
I think the starting point would be to use the uncorrected DOI string; "ihaledeposotedinthecopnttol..."

And then have AI separate and correct the message.
"I have deposited in the county of.."

Once AI can do that, then the next step would be to bring in a "text" document to remove punctuation, headers, the spaces, etc. Documents before 1820 puts the first word of the next at the bottom. That too needs to be removed.

Then do the same for different formats: pdfs, jpgs, etc. There are probably converter to do this that could be leveraged: pdf to text, jpg to text, etc.

ihale vs ihave
Number 807 is correct - VALUABLE. It is good text: I have...
Just take DOI from a pamphlet.

 

[chalkoutline]

ihale vs ihave
Number 807 is correct - VALUABLE. It is good text: I have...
Just take DOI from a pamphlet.

You are correct; 807 is “Valuable” and used to spell “have”; However, when you use the numbered DOI from the 1885 pamphlet, the word “Valuable” ends up being 818; not 807. This is because the pamphlet used a different DOI. Also, when TB numbered the DOI, he skipped several words causing words in the decoded message to be misspelled. If I remember correctly, you don’t see correctly spelled words until you get to “FOUR MILES” from Bufords.

My post above suggests the use of AI to look for correctly spelled words and add a space before and after, then have AI look for misspelled words and correct them…correct “hale” to “have”.

This would eliminate the tedious process of finding where and how many words to skip.

 

[chalkoutline]

Within another thread, Advanceforgold suggested using AI to solve the ciphers which makes a lot of sense.so started a new thread for this topic.

Advanceforgold also suggested using python.

I've mentioned my Beale Decoder needs a text version. To extract the text from a document image, you can use:

• • Windows Snipping Tool:
    Capture a screenshot of the image, use the snipping tool to select the text area, and copy the extracted text.

You then need to remove all punctuation and spaces.

This is another use for AI.

 

[Juskevicius]

>>However, when you use the numbered DOI from the 1885 pamphlet, the word “Valuable” ends up being 818; not 807.
No. Look in 1885 pamphlet. “Valuable” is number 807.

 

[chalkoutline]

>>However, when you use the numbered DOI from the 1885 pamphlet, the word “Valuable” ends up being 818; not 807.
No. Look in 1885 pamphlet. “Valuable” is number 807.

Your pamphlet appears to be different than mine. I didn't want to retype the DOI so I found a pamphlet that was already in text form. I attached a screenshot for the area we've been discussing.

 

[chalkoutline]

Your pamphlet appears to be different than mine. I didn't want to retype the DOI so I found a pamphlet that was already in text form. I attached a screenshot for the area we've been discussing.

I guess that means...if you are going to use AI to search all docs, books, newspaper s, etc., you need to include all versions and reprints. 😁