What Single-Topic Summaries Miss in Hospital Reviews—Aspect-Level Evaluative Structure Using Generative Pretrained Transformer–Based Sentiment Analysis: Content Analysis
Jung-Tang Hsueh, Sheng-Hsun Hsu, Shwu-Fen Chiu · 2026
Read What Single-Topic Summaries Miss in Hospital Reviews—Aspect-Level Evaluative Structure Using Generative Pretrained Transformer–Based Sentiment Analysis: Content Analysis by Jung-Tang Hsueh, Sheng-Hsun Hsu, Shwu-Fen Chiu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13620426.txt.
Description
Jung-Tang Hsueh, Sheng-Hsun Hsu, and Shwu-Fen Chiu published a text analysis study in two thousand twenty-six on online hospital reviews. Standard methods tag each review with one topic, yet patients mix praise and complaint freely. That flattening may hide how services really perform.
The team studied over five thousand four hundred Google reviews of all twenty-four medical centres in Taiwan from two thousand twenty-four. They compared the classic single-topic method with a generative AI approach scoring seven service aspects separately. Human raters agreed strongly, and the AI matched the consensus closely. Mixed-feeling reviews proved common. Positive and negative reviews also differed in which aspects clustered together. The authors say aspect-level scoring preserves what one-topic summaries throw away.
Catalog record
| CREATOR | Jung-Tang Hsueh, Sheng-Hsun Hsu, Shwu-Fen Chiu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |