What data looks like for this category
This category of data covers daily review documents including infectious disease report cards, epidemic analysis reports, case correction records, clustered epidemic reports, and epidemiological investigation reports. Data sources include the Chinese Center for Disease Control and Prevention Information System, hospital HIS and infectious disease reporting systems, and laboratory testing systems. Updates occur daily. Document structures include fields such as basic case information, diagnosis results, reporting time limits, correction notes, and test results. Some complex review scenarios require cross-verified data from multiple systems.
What constraints do these characteristics impose on the citation sources and traceability link
Daily updated data sources require the citation link to support real-time synchronization of the latest case data, to avoid using expired review bases. Multi-system cross data sources require the traceability link to clearly mark the source system of each cited content, to ensure that the complete data link can be traced during review. Documents include mandatory fields for statutory infectious disease reports. When citing, it is necessary to accurately match the associated content of the corresponding fields, to avoid incorrect cross-field citations. At the same time, file review requires retaining the complete citation link to meet traceability verification requirements for subsequent supervision and inspections.
How to set configurations
Reference only; verify against the target FastGPT release and deployment documentation.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
quoteMaxToken | 800–1200 token | Matches the core content length of a single infectious disease report card or epidemiological investigation report, ensuring complete citation of review basis |
responseData | Return only associated data sources | Limit output to cited content with clearly marked sources, meeting traceability requirements for disease control review |
detail | Enable full traceability | Display the original source path and field matching relationship of cited content, facilitating cross-verification for file review |
| Maximum context | 15000–20000 characters | Accommodate aggregated content from multiple associated documents, adapting to daily updated batch review scenarios |
| Single search limit | Top 6 entries | Match the daily reported case volume range, avoiding redundant citations that interfere with review judgments |
Three common configuration errors
- Phenomenon: When configuring citation parameters, only preset options of 100 or 900 are available, and custom ranges around 300 cannot be set. Cause: Parameter custom permission is not enabled, and only the platform's preset fixed configuration levels are used.
- Phenomenon: Unmarked non-official data is mixed into cited content. Cause: The source filtering function of
responseDatais not enabled, resulting in irrelevant content from non-designated systems being introduced. - Phenomenon: It is impossible to trace back to the original document of a case correction record during review. Cause: The full traceability configuration of
detailis not enabled, and the original system data link of the correction record is not associated.
How to confirm configurations are correctly set
- A test infectious disease report card is uploaded, and cited content is checked to confirm it only includes document fragments from designated source systems.
- Markings on each cited content are reviewed to confirm all have clear data source identifiers and field corresponding relationships.
- Customization of the
quoteMaxTokenvalue is attempted, and verification is conducted to confirm adjustment is possible without relying on preset options. - A file review scenario is simulated, and verification is conducted to confirm the complete path to the original case correction record or epidemiological investigation report can be traced back.
The values provided here are common starting points and should be evaluated against applicable sample datasets.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.