What this category's data looks like
Materials for business basis queries at peer review organizations typically include three types: review invitations, expert review comments, and author revision notes. A review invitation contains the manuscript number, submission date, expert name, and invitation receipt; expert review comments contain the review conclusion, comment body, submission date, and signature; author revision notes contain point-by-point responses, the revised manuscript version number, and the submission date. Materials are continuously updated by the editorial office by round; the same manuscript produces multiple files across different rounds, and fields change with the round. After entering this feature, the three types of materials are each parsed and chunked as independent file records, and stored as data rows bound to a unique data identifier, for later retrieval to trace back to the original text.
Configuration and implementation checks
Parsing and training tasks (server-side tasks) only output data rows and index status (parsed, indexing, indexed); basis entries that can trace back to the original text must be handled by retrieval and subsequent processing. After content parsing is complete, parsing and chunking are immediately stored as visible data rows; indexing flows through three states, and data pending indexing does not participate in recall.
| Object | Check | Verification action |
|---|---|---|
| Review invitation | Fields include the manuscript number, submission date, expert name, and invitation receipt; when the receipt is missing, the basis entry cannot be mapped to a specific expert | Check whether the identifiers of the file record and the data row are distinguishable, and confirm that the invitation receipt field has a value |
| Expert review comments | Fields include the review conclusion, comment body, submission date, and signature; the names, quantities, and dates of the same manuscript in the two materials must match up | Open the data row and check whether the status is parsed / indexing / indexed, and compare the dates and signatures one by one |
| Author revision notes | Fields include point-by-point responses, the revised manuscript version number, and the submission date; when the version is not synchronized, the responses do not match the current materials | Check the file record and the data row separately, and confirm that the version number is consistent with the current materials |
| Data row | Is in one of the three states and is consistent with the interface display; the rebuild scope covers only data that has completed indexing | Look at the data rows covered by the rebuild task, and check the records of trainingType and chunkSize, chunkSplitter |
Three common errors
- Symptom: the names, quantities, and dates of the same manuscript number do not match between the review invitation and the expert review comments. Cause: the two materials are parsed and stored separately, lacking cross-material comparison.
- Symptom: the version number of the author revision notes is inconsistent with the current materials, and the response entries point to an old manuscript. Cause: the version is not synchronized, and the old data row is still in the completed indexing state.
- Symptom: the review invitation lacks an invitation receipt, and the basis entry cannot trace back to a specific expert. Cause: the signature or receipt field is empty, and the data row is still indexed after parsing.
How to verify the setup
Feature acceptance: upload one material, observe the data row status transitions and the coverage scope of the rebuild task, and confirm that only parsed / indexing / indexed states appear, and that the rebuild covers only data rows that have completed indexing.
Business test: ask a cross-material basis question, and see whether the answer provides a traceable source, and whether the source matches the original entries in the review invitation, expert review comments, or author revision notes.
- Open the data row and check the status, confirm it is consistent with the interface display, and check the file record and the data row separately
- Look at the data rows covered by the rebuild task, and check the records of
trainingType,chunkSize,chunkSplitter