TY - JOUR
T1 - Bidirectional GRU for targeted aspect-based sentiment analysis based on character-enhanced token-embedding and multi-level attention
AU - Setiawan, Esther Irawati
AU - Ferry, Ferry
AU - Santoso, Joan
AU - Sumpeno, Surya
AU - Fujisawa, Kimiya
AU - Purnomo, Mauridhi Hery
N1 - Publisher Copyright:
© 2020, Intelligent Network and Systems Society.
PY - 2020/10/1
Y1 - 2020/10/1
N2 - The user's feedback on healthcare services is usually based on ratings from post-service questionnaires. However, in order to get a clear view of the user's perspective, online text reviews need to be analyzed. We combined targeted and aspect-based sentiment analysis by multi-level attention to get a specific user sentiment on a target of an aspect. The multi-level attention consists of Target-level and Sentence-level attention. Our proposed framework is based on Bidirectional Gated Recurrent Unit. Bi-GRU is commonly known to have comparable results compared to LSTM while having lesser computational complexity. We also utilized Bidirectional LSTM based Character-Enhanced Token-Embedding to handle out of vocabulary words and misspelling to avoid error in detecting sentiment. We created a dataset of online healthcare reviews from 2018-2020, targeting the name of the hospital or department, with ten aspects: cleanliness, cost, doctor, food, nurse, parking, receptionist and billing, safety, test and examination, and waiting time. To improve the results of our proposed method, we calculated polarity weight to handle imbalanced aspects in the dataset. We classified these reviews into three polarities, which are positive, negative, and neutral. Based on our experiments, we achieved the best F1-Score of 88%.
AB - The user's feedback on healthcare services is usually based on ratings from post-service questionnaires. However, in order to get a clear view of the user's perspective, online text reviews need to be analyzed. We combined targeted and aspect-based sentiment analysis by multi-level attention to get a specific user sentiment on a target of an aspect. The multi-level attention consists of Target-level and Sentence-level attention. Our proposed framework is based on Bidirectional Gated Recurrent Unit. Bi-GRU is commonly known to have comparable results compared to LSTM while having lesser computational complexity. We also utilized Bidirectional LSTM based Character-Enhanced Token-Embedding to handle out of vocabulary words and misspelling to avoid error in detecting sentiment. We created a dataset of online healthcare reviews from 2018-2020, targeting the name of the hospital or department, with ten aspects: cleanliness, cost, doctor, food, nurse, parking, receptionist and billing, safety, test and examination, and waiting time. To improve the results of our proposed method, we calculated polarity weight to handle imbalanced aspects in the dataset. We classified these reviews into three polarities, which are positive, negative, and neutral. Based on our experiments, we achieved the best F1-Score of 88%.
KW - Bidirectional GRU
KW - Character-enhanced token-embedding
KW - Multi-level attention
KW - Targeted aspect-based sentiment analysis
UR - http://www.scopus.com/inward/record.url?scp=85090390240&partnerID=8YFLogxK
U2 - 10.22266/ijies2020.1031.35
DO - 10.22266/ijies2020.1031.35
M3 - Article
AN - SCOPUS:85090390240
SN - 2185-310X
VL - 13
SP - 392
EP - 407
JO - International Journal of Intelligent Engineering and Systems
JF - International Journal of Intelligent Engineering and Systems
IS - 5
ER -