|
Abstract
|
The improvement of service qualities of online medical communities is closely related to the service needs information of patients, therefore, this study adopts the BERTopic model to mine the differences in the service needs of patients with physiological and psychological diseases, which provides a basis for decision-making on the targeted improvement of service qualities of online medical communities. Post-visit comments of cardiovascular medicine and psychiatric-psychological patients were collected from an online medical community using Octopus. After data preprocessing, we constructed a BERTopic model to identify topics related to the different needs of patients with physiological and psychological diseases. In addition, the sentiment dictionary for the domains of physiological and psychological diseases was expanded by the SOPMI algorithm, and the average sentiment scores were calculated for the comments under each topic of different diseases using sentiment analysis methods. It was found that patients with psychological diseases had stronger sentimental tendencies and more prominent service needs than patients with physical diseases. The difference in service needs between patients with physiological and psychological diseases is mainly reflected in the diagnostic examination, treatment plan, treatment effect, and service attitude. Therefore, the quality of medical services should be improved according to the different needs of patients with different types of diseases in terms of diagnosis and examination, treatment plan, treatment effect, and service attitude.
|