... is the process of matching different language lexicons. Can be accomplished by word-counting algorithms in source and destination texts. The algorithm is run separately on the source and destination texts whereby each unique word is assigned a numerical word-frequency value. In this way, each text can be divided into low-frequency and high-frequency words.
Presumably the low-frequency words in the source text can be matched with the low-frequency words in the destination text and provide a translation dictionary. These single- or multi-word strings form associations which a language speaker would normally make in each sentence.
Visualised as follows:
Preceding1 association1 following1/preceding2 association2 following2.
following1=preceding2
... further explanation to be added.
