TY - GEN
T1 - Semantic loss in autoencoder tree reconstruction based on different tuple-based algorithms
AU - Calvo, Hiram
AU - Rivera-Camacho, Ramón
AU - Barrón-Fernndez, Ricardo
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2018.
PY - 2018
Y1 - 2018
N2 - Current natural language processing analysis is mainly based on two different kinds of representation: structured data or word embeddings (WE). Modern applications also develop some kind of processing after based on these latter representations. Several works choose to structure data by building WE-based semantic trees that hold the maximum amount of semantic information. Many different approaches have been explores, but only a few comparisons have been performed. In this work we developed a compatible tuple base representation for Stanford dependency trees that allows us to compared two different ways of constructing tuples. Our measures mainly comprise tree reconstruction error, mean error over batches of given trees and performance on training stage.
AB - Current natural language processing analysis is mainly based on two different kinds of representation: structured data or word embeddings (WE). Modern applications also develop some kind of processing after based on these latter representations. Several works choose to structure data by building WE-based semantic trees that hold the maximum amount of semantic information. Many different approaches have been explores, but only a few comparisons have been performed. In this work we developed a compatible tuple base representation for Stanford dependency trees that allows us to compared two different ways of constructing tuples. Our measures mainly comprise tree reconstruction error, mean error over batches of given trees and performance on training stage.
KW - Parsing
KW - Semantic reconstruction
KW - Structuring word embeddings
UR - http://www.scopus.com/inward/record.url?scp=85057286143&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-01132-1_20
DO - 10.1007/978-3-030-01132-1_20
M3 - Contribución a la conferencia
SN - 9783030011314
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 174
EP - 181
BT - Progress in Artificial Intelligence and Pattern Recognition - 6th International Workshop, IWAIPR 2018, Proceedings
A2 - Heredia, Yanio Hernández
A2 - Núñez, Vladimir Milián
A2 - Shulcloper, José Ruiz
PB - Springer Verlag
T2 - 6th International Workshop on Artificial Intelligence and Pattern Recognition, IWAIPR 2018
Y2 - 24 September 2018 through 26 September 2018
ER -