ololololo dateline='[url=tel:1795247925' wrote: You are not allowed to view links. Register or Login to view.1795247925[/url]']
A bit simpler.
If a glyph is most often the first letter in a word (like ch), it is given a high weight. If it is mainly only at the end (like iin and m), it is given a low weight.
First, these weights can be used to conditionally determine which symbol is “larger” and “heavier.” For example, ch is heavier than a, but a is heavier than iin and r. The same applies to bigrams: ch is heavier than or, but or is heavier than a.
Secondly, with this help, it will be possible to represent words as a sequence of digits. For example, daiin = 321.
oh I get it so your hypothesis is that bigrams that appear most frequently in the beginning of a word have higher weight / numerical value and so on.
but how would we map this consistently if the same bigram appears in the middle / end of a word as well. This would make sense only if it was preceded by a higher weight bigram consistently.
here’s an example, “or” appears 574/2591=0.222 in line starts, 297 /2591=0.115 middles and 1720/2591=0.664 line ends.
hence it should always be preceded by a bigram with higher weight and followed by a bigram with lower weight.
but sometimes it gets doubled like “oror” and sometimes it is preceded and followed by the same bigram like following.
preceded: ol-or , counts: 63
followed: or-ol, counts: 38
how would you interpret this in the context of the hypothesis?

