Vowel identification - Sukhotin's Algorithm

Welcome to Crypto. We hope you enjoy your visit.

You're currently viewing our forum as a guest. This means you are limited to certain areas of the board and there are some features you can't use. If you join our community, you'll be able to access member-only sections, and use many member-only features such as customizing your profile, sending personal messages, and voting in polls. Registration is simple, fast, and completely free.

Join our community!

If you're already a member please log in to your account to access all of our features:

Vowel identification - Sukhotin's Algorithm
Topic Started: Oct 29 2008, 03:34 AM (180 Views)
jdege	Oct 29 2008, 03:34 AM Post #1
Elite member Posts: 263 Group: Admin Member #246 Joined: December 7, 2006	The November-December 2008 issue of "The Cryptogram" arrived the other day. PARROT has an article "Solving P-Sp-1". In it, he references an algorithm for identifying vowels I'd not heard of before. That sent me on a bit of a research spree. B.V. Sukhotin published it in a Russian linguistics journal, back in 1962. A translation was published in a French information systems journal in 1973. The first English language description was in 1991, by Jacques Guy, in Cryptologia 15:3,258-262. It was later discussed by PHEONIX in The Cryptogram MA92 and SO92. Anyone heard of this? The Cm SO92 article measures its effectiveness compared to a number of other published vowel-identification techniques, and ranks it the most effective. Yet I've never seen it discussed before. A Google search finds hits on a 1992 Cryptologia article by Caxton Foster that does similar comparisons with the same results. (Was PHOENIX Caxton Foster? The nym is no longer in the ACA index.) The process is simple. Code: 1.[space]Create[space]a[space]26[space]x[space]26[space]matrix[space]of[space]the[space]number[space]of[space]times[space]each letter[space]contacts[space]another.[space]The[space]result[space]will[space]be[space]diagonally[space]symmetrical. 2.[space]Set[space]all[space]the[space]cells[space]on[space]the[space]main[space]diagonal[space]to[space]zero. 3.[space]Create[space]a[space]1[space]x[space]26[space]vector[space]of[space]the[space]sums[space]of[space]the[space]rows. 4.[space]Mark[space]the[space]highest[space]sum[space]that[space]is[space]greater[space]than[space]zero[space]as[space]a[space]vowel. 5.[space]From[space]each[space]row[space]sum[space]that[space]is[space]not[space]marked[space]as[space]a[space]vowel[space]subtract twice[space]the[space]number[space]of[space]times[space]that[space]the[space]row[space]letter[space]occurs[space]next[space]to your[space]new[space]vowel. 6.[space]If[space]there[space]remain[space]non-vowel[space]rows[space]with[space]sums[space]greater[space]than[space]zero,[space] go[space]to[space]step[space]4. 7.[space]Exit. Simple enough. It only took me a couple of minutes to write an awk script that would implement it. It does work. But despite the articles on how well it works compared to other methods, it doesn't seem to work as well as the method I described in Another technique for identifying vowels What is different, though, about this method is that it can be done with paper and pencil.
	When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.


Revelation	Oct 29 2008, 06:47 PM Post #2
Administrator Posts: 386 Group: Admin Member #1 Joined: August 23, 2005	I've received the issue too. This is interesting stuff!
	RRRREJMEEEEEPVKLWENFNVJKEEEEEAOLKAFKLXCFZAASDJXZTTTTTTTLSIOWJXMOKLAFJNNKFNXN RAGRBAQEMHIGDJVDSEOXVIYCELFHWLELJFIENXLRATALSJFSLCYTKLASJDKMHGOVOKAJDNMNUITN RRRRLJVEEEEECLYVYHNVPFTAEEEEEMWLMEIRNGLARWJAKJDFLWNTIERJMIPQWOTZEOCXKNUBNXCN RJIRPOWEANFUSNCZVDVZNMSFEKLOEPZLDKDJWSAAAAAAAOERHJCTNCKFRIMVKSOFOMKMANREWNBN RZUDRGXEEEEENFQIDVLQNCKNEEEEEDGLLLLLLAWIOSNCDARLODMTOEJXMILDFJROTKJSDNLVCZNN


jdege	Oct 29 2008, 07:02 PM Post #3
Elite member Posts: 263 Group: Admin Member #246 Joined: December 7, 2006	I loved how he started out with an incorrect assumption, and from it managed to work his way to a solution. It's the way it usually happens for me. (Haven't started on any of the Cons, yet. Waiting for the digital copy. Hate wasting my time working around my typos.
	When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.


Revelation	Oct 29 2008, 11:10 PM Post #4
Administrator Posts: 386 Group: Admin Member #1 Joined: August 23, 2005	I've made an implementation too, in Python. Be warned: this is my very first python program. It took me a while to find out how to set the size of an array (sum = [0] * 26). Can't believe that that's not in the documentation. Attached to this post: vowel.py (926 Bytes) Edited by Revelation, Oct 29 2008, 11:14 PM.
	RRRREJMEEEEEPVKLWENFNVJKEEEEEAOLKAFKLXCFZAASDJXZTTTTTTTLSIOWJXMOKLAFJNNKFNXN RAGRBAQEMHIGDJVDSEOXVIYCELFHWLELJFIENXLRATALSJFSLCYTKLASJDKMHGOVOKAJDNMNUITN RRRRLJVEEEEECLYVYHNVPFTAEEEEEMWLMEIRNGLARWJAKJDFLWNTIERJMIPQWOTZEOCXKNUBNXCN RJIRPOWEANFUSNCZVDVZNMSFEKLOEPZLDKDJWSAAAAAAAOERHJCTNCKFRIMVKSOFOMKMANREWNBN RZUDRGXEEEEENFQIDVLQNCKNEEEEEDGLLLLLLAWIOSNCDARLODMTOEJXMILDFJROTKJSDNLVCZNN


jdege	Oct 30 2008, 12:31 AM Post #5
Elite member Posts: 263 Group: Admin Member #246 Joined: December 7, 2006	One of the articles in Cm suggests that instead of calculating the sum for each row, you calculate the sum and the variety of contact (VOC). Then choose as the vowel the largest product of sum and VOC. Then decrement the sum by 2x the count, as we had been doing, and decrement the VOC by 1x the count. If any sum goes below zero, set it to 0, otherwise you could see a negative sum and a negative VOC be the highest product.
	When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.


1 user reading this topic (1 Guest and 0 Anonymous)


« Previous Topic · General · Next Topic »