2
$\begingroup$

Is there any general idea on how humans solve jumbled words? I know many people will say we match it against a commonly used words checklist mentally, but it is kind of vague. Is there any theory on this? And how might an AI learn to do the same?

$\endgroup$
0

1 Answer 1

4
$\begingroup$

I remember a problem similar to this https://vladris.com/puzzles/facebook/puzzle_master/snack/breathalyzer.html this is given as a problem in facebook engineering puzzles.

But i doubt that it is not much of a machine learning/AI problem. you could implement an algorithm that converts each word into a set of its characters , then pick the word in your master list with minimum distance based on its character list .

Even when humans solve jumble we do it in a systematic/algorithmic way , if the jumbled word is not present in our memory we can't do anything otherwise we can solve the jumble.

But human brain can simply recognize scrabled words if some of the structure is retained and not fully scrambled like an anagram .

We generally index words based on the first syllable/character in our brain , like enter image description here or even this enter image description here

In the second picture we recognised the words even if they are not made of english characters because our brain doesn't scan the text character by character like "7-H " but treats it like an image.

So the model should not immediately classify the segmented characters but should find the "nearest" characters in every class that optimises the combined probability of characters of the word being one of the classes of words we have in our dictionary.

$\endgroup$
4
  • $\begingroup$ Great answer..But in the first example if you notice, all the complex words are kept with their starting letter untouched..Even for machines finding nearest distance can be a tough job speaking from purely combinatorics viewpoint..As far as I solve I just try to speak different combinations of the letter mentally and suddenly it matches with pre heard sound wave and we have a match $\endgroup$
    – user9947
    Commented Jun 30, 2018 at 18:29
  • 1
    $\begingroup$ for making the model more human-like which by the way suits the first example , you could include an "attention" mechanism that focuses more on the parts of the words which humans emphasize more when reading $\endgroup$ Commented Jun 30, 2018 at 18:30
  • $\begingroup$ Like vowels I guess $\endgroup$
    – user9947
    Commented Jun 30, 2018 at 18:32
  • 1
    $\begingroup$ or it could be framed as a reinforcement learning problem , where the reward is +1 when the solved word is in the dictionary like in arxiv.org/pdf/1805.07470.pdf $\endgroup$ Commented Jun 30, 2018 at 18:36

You must log in to answer this question.