Construction of an aspect-level sentiment analysis model for online medical reviews

Online medical services have become increasingly popular, and patient feedback can significantly influence other patients' medical decision-making. This study utilizes a double-layer domain ontology for conducting aspect-level sentiment analysis of reviews from online medical platforms. A double-layer aspect recognition model (OMR-ARM), aggregating the knowledge of the domain ontology, is built to identify the aspects of online medical reviews. The proposed model outperforms baseline models by up to 23.12%. Incorporating this model into a series of state-of-theart models, the resultant OMR-ALSA model achieves a F1-score of 93.53% for aspect-level sentiment analysis of online medical reviews. Additionally, this study develops an objectaspect-sentiment knowledge graph of online medical reviews (OMR-KG) that can classify patients' sentimental polarities towards the different aspects of online medical reviews. The proposed model and constructed KG have the potential to provide reference and guidance to sentiment analysis research in the online medical review domain, thus contributing to more informed and personalized healthcare decision-making.

Automated summary

This study utilizes a double-layer domain ontology for aspect-level sentiment analysis of online medical reviews. A double-layer aspect recognition model is built, and an object-aspect-sentiment knowledge graph is constructed, providing reference and guidance to sentiment analysis research in the online medical review domain.