09-08-2020, 05:17 AM
A hypervector was generated for each language. They are shown compared on the plot. You didn't read the linked article. I wrote:
Now we apply Kanerva’s theory. Assign random hypervectors to the twenty six letters of the alphabet and a space character. Parse any large text sample with a sliding three letter window, building a series of trigrams. Combining the trigrams yields a hyperdimensional vector signature for the text sample. Languages which are similar will have hypervectors close together. (Like the previous V and W vectors) Dissimilar languages will have orthogonal vectors.
Now we apply Kanerva’s theory. Assign random hypervectors to the twenty six letters of the alphabet and a space character. Parse any large text sample with a sliding three letter window, building a series of trigrams. Combining the trigrams yields a hyperdimensional vector signature for the text sample. Languages which are similar will have hypervectors close together. (Like the previous V and W vectors) Dissimilar languages will have orthogonal vectors.