TY - CHAP
T1 - Syntactic n-grams
T2 - The concept
AU - Sidorov, Grigori
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2019.
PY - 2019
Y1 - 2019
N2 - As we have already mentioned, the main idea of the formal features applicable in computational linguistics is related to the vector space model and the use of n-grams as features in this space, which also includes unigrams, i.e., words. The words are considered in the contexts. Usually, it is neighbor words. But some words that have syntactic relations are not neighbors, thus, here appears the idea to use syntactic information to obtain real (syntactically related) neigbors as the context of a word. So, we suggest obtaining n-grams by following paths in sytanctic trees.
AB - As we have already mentioned, the main idea of the formal features applicable in computational linguistics is related to the vector space model and the use of n-grams as features in this space, which also includes unigrams, i.e., words. The words are considered in the contexts. Usually, it is neighbor words. But some words that have syntactic relations are not neighbors, thus, here appears the idea to use syntactic information to obtain real (syntactically related) neigbors as the context of a word. So, we suggest obtaining n-grams by following paths in sytanctic trees.
UR - http://www.scopus.com/inward/record.url?scp=85064594687&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-14771-6_8
DO - 10.1007/978-3-030-14771-6_8
M3 - Capítulo
AN - SCOPUS:85064594687
T3 - SpringerBriefs in Computer Science
SP - 47
EP - 58
BT - SpringerBriefs in Computer Science
PB - Springer
ER -