TY - JOUR
T1 - Improving aspect-level sentiment analysis with aspect extraction
AU - Majumder, Navonil
AU - Bhardwaj, Rishabh
AU - Poria, Soujanya
AU - Gelbukh, Alexander
AU - Hussain, Amir
N1 - Publisher Copyright:
© 2020, Springer-Verlag London Ltd., part of Springer Nature.
PY - 2022/6
Y1 - 2022/6
N2 - Aspect-based sentiment analysis (ABSA), a popular research area in NLP, has two distinct parts—aspect extraction (AE) and labelling the aspects with sentiment polarity (ALSA). Although distinct, these two tasks are highly correlated. The work primarily hypothesizes that transferring knowledge from a pre-trained AE model can benefit the performance of ALSA models. Based on this hypothesis, word embeddings are obtained during AE and, subsequently, feed that to the ALSA model. Empirically, this work shows that the added information significantly improves the performance of three different baseline ALSA models on two distinct domains. This improvement also translates well across domains between AE and ALSA tasks.
AB - Aspect-based sentiment analysis (ABSA), a popular research area in NLP, has two distinct parts—aspect extraction (AE) and labelling the aspects with sentiment polarity (ALSA). Although distinct, these two tasks are highly correlated. The work primarily hypothesizes that transferring knowledge from a pre-trained AE model can benefit the performance of ALSA models. Based on this hypothesis, word embeddings are obtained during AE and, subsequently, feed that to the ALSA model. Empirically, this work shows that the added information significantly improves the performance of three different baseline ALSA models on two distinct domains. This improvement also translates well across domains between AE and ALSA tasks.
KW - AE
KW - ALSA
KW - Knowledge transfer
UR - http://www.scopus.com/inward/record.url?scp=85089469609&partnerID=8YFLogxK
U2 - 10.1007/s00521-020-05287-7
DO - 10.1007/s00521-020-05287-7
M3 - Artículo
AN - SCOPUS:85089469609
SN - 0941-0643
VL - 34
SP - 8333
EP - 8343
JO - Neural Computing and Applications
JF - Neural Computing and Applications
IS - 11
ER -