Mapping Clinical Intent in Internet Hospitals: Identifying and Profiling Patient Demand via Natural Language Processing
Ruoxin Xu, Jing Liu, Huizi Ye, Tingting Yang, Kuanlie Jiang, Xiaojiao Zhu · 2026
Read Mapping Clinical Intent in Internet Hospitals: Identifying and Profiling Patient Demand via Natural Language Processing by Ruoxin Xu, Jing Liu, Huizi Ye, Tingting Yang, Kuanlie Jiang, Xiaojiao Zhu (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-13632933.txt.
Description
What patients actually want when they message an internet hospital is decoded by Xu and colleagues. They used language tools on real enquiries. Rigid online menus often mismatch what patients mean. That wastes triage time and sends resources to the wrong place. The team analysed more than twenty-eight thousand messages from a large Chinese internet hospital, first clustering them without labels to discover intent types, then training a specialised language model to classify them.
Clustering revealed eight distinct intents, with clear overlap between rehab and follow-up requests. The tuned small model beat two large language models on a gold-standard set of marked records. It also held up better on a separate nationwide dataset. Department patterns differed sharply, since prescription renewals dominated neurology enquiries while first symptom questions dominated dermatology, giving hospitals a measurable basis for redesigning digital services.
Catalog record
| CREATOR | Ruoxin Xu, Jing Liu, Huizi Ye, Tingting Yang, Kuanlie Jiang, Xiaojiao Zhu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |