Syntagmatic relations correspond to relations of contiguity that words maintain within a given syntactic context (sentence, chunk, fixed length window, etc)., two words being related if they often appear together.
Reading the 3bn "base pairs" in human DNA - akin to letters, encoding a total of between 20,000 and 30,000 genes that are the "words" of genetics - is getting faster as companies find quicker ways to "read" entire stretches of DNA at a time, like reading a sentence in chunks rather than letter by letter.
They first segment the document into units like sections, sentences, add part-of-speech tags, and split sentences into chunks, especially noun phrases.
Obviously, the sentence was chunked into single words, even syllables, associated with underlying accent commands and often separated by pauses.
At the same time, focus on basis sentences, ready made chunks, such as how are you, my name is, how old are you, etc.
In cases where a sentence or a chunk of text could be analyzed as serving more than one rhetorical function (move or step), the conventional practice in the field is to assign the most salient move or step (Crookes 1986; Holmes 1997).
In this manner we applied a pipeline of rudimentary operations which included word stemming, sentence detection and chunking.
Most of the errors are related to complementary processing such as inaccurate sentence boundary identification, chunking, Part-Of-Speech (POS) tagging or word sense disambiguation in the Meta-thesaurus.
For example, " [The/DT anti-androgenic/JJ activity/NN] of/IN [the/DT ethanol/NN extract/NN] of/IN [the/DT fruiting/VBG body/NN] of/IN [Ganoderma/NN lucidum/NN] has/VBZ been/VBN previously/RB reported/VBN./." is a POS-tagged and NP chunked sentence where each square bracket represents a NP.
Plus it's a pretty small working area so you're probably going to want to translate in chunks, a sentence at a time, say.
