Reviews Analysis of Korean Clinics Using LDA Topic Modeling

Objectives: In the health care industry, influence of online reviews is growing. As medical services are provided mainly by providers, those have been managed hospitals and clinics. However, direct promotions providers legally forbidden. Due to this reason, consumers, like patients clients, search a lot on Internet get any information about hospitals, treatments, prices, etc. It can be determined that indicate quality analysis should done for sustainable hospital marketing.Method: Using Python-based crawler, we collected reviews, written real patients, who had experienced Korean medicine, more than 14,000 reviews. To extract most representative words, were divided positive negative; after pre-processed only nouns adjectives TF(Term Frequency), DF(Document TF-IDF(Term Frequency – Inverse Document Frequency). Finally, some topics aggregations extracted words analyzed using LDA(Latent Dirichlet Allocation) methods. avoid overlap, number set Davis visualization.Results Conclusions: 6 3 in each positive/negative review, LDA Topic Model. The main factors, consisting 1) Response customers. 2) Customized treatment (consultation) management. 3) Hospital/Clinic’s environments.