SA2SL: From Aspect-Based Sentiment Analysis to Social Listening System for Business Intelligence

In this paper, we present a process of building a social listening system\nbased on aspect-based sentiment analysis in Vietnamese from creating a dataset\nto building a real application. Firstly, we create UIT-ViSFD, a Vietnamese\nSmartphone Feedback Dataset as a new benchmark corpus built based on a strict\nannotation schemes for evaluating aspect-based sentiment analysis, consisting\nof 11,122 human-annotated comments for mobile e-commerce, which is freely\navailable for research purposes. We also present a proposed approach based on\nthe Bi-LSTM architecture with the fastText word embeddings for the Vietnamese\naspect based sentiment task. Our experiments show that our approach achieves\nthe best performances with the F1-score of 84.48% for the aspect task and\n63.06% for the sentiment task, which performs several conventional machine\nlearning and deep learning systems. Last but not least, we build SA2SL, a\nsocial listening system based on the best performance model on our dataset,\nwhich will inspire more social listening systems in future.\n