What this category's data looks like
This category's data covers three types of materials: program setup application forms, talent cultivation plans, and expert review opinions. The sources are academic affairs and departmental submission archives, and the update cadence changes with submission batches, review meetings, and approval milestones. The three types of materials each have their own focus. The application form contains the program name, code, program duration, enrollment scale, and responsible person's signature; the cultivation plan contains course modules, credits and credit hours, practical components, and the approval date; the review opinions contain the expert list, review conclusion, date, and signature receipt. When the materials enter the document parsing and chunking function, each is stored as an independent data row, bound to a unique data identifier, and becomes visible after parsing and chunking are complete, providing a traceable data foundation for subsequent retrieval and version ledgers.
Configuration and implementation checks
This step is executed by the parsing and training task (server-side task), which outputs data rows and index status (parsed, indexing, indexed). The "version ledger and change records" are then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Program setup application form | The name, code, program duration, enrollment scale, and responsible person's signature must be consistent with the talent cultivation plan | Check the program name, code, program duration, and dates in the application form and cultivation plan, and confirm that signatures and receipts are complete; check whether the data row status is parsed or indexed |
| Talent cultivation plan | The course modules, credits and credit hours, practical components, and approval date must connect with the review opinions | Check the number of courses, total credits, and approval date in the cultivation plan and expert review opinions, and confirm that the old version can be traced when versions are not synchronized; check whether the chunking results of chunkSize and chunkSplitter preserve field boundaries |
| Expert review opinions | The expert list, review conclusion, date, and signature receipt must be complete and correspond to the application form | Check the number of experts, conclusion date, and application form date in the review opinions, and confirm that missing signatures or receipts are marked as pending supplementation; check whether trainingType and the three-state index flow are consistent |
| Data rows and file records | File records and data rows each have separate identities, and status is consistent with the interface display | Check file records and data rows separately, and confirm that data pending indexing does not participate in recall; rebuild tasks only cover data rows whose indexing has been completed |
Three common errors
- Symptom: the same program name is written differently in the application form and the cultivation plan. Cause: the two materials are filled in by different roles, with no field alignment or version synchronization, so recall results are inconsistent during retrieval.
- Symptom: the number of courses and total credits do not match between the two materials. Cause: after the cultivation plan was revised, it was not written back to the application form, the old version was not marked invalid, and the version ledger lacks change records.
- Symptom: expert review opinions lack signatures or receipts but still enter the index. Cause: parsing and chunking only split content and do not validate signature fields, and the index status does not block materials pending supplementation.
How to verify the setup
This feature's acceptance: upload one material and observe the data row status flow and the coverage scope of the rebuild task. The status should go from parsed to indexing to indexed, and data pending indexing does not participate in recall; rebuilding with a different model only covers indexed data, so check trainingType and the data rows covered by the rebuild task.
Business test: change one piece of content and see whether the version number and references are updated in sync. After modifying the credits in the cultivation plan, confirm whether the references in the application form and review opinions point to the new version and whether the old version can be traced.
- Open the data row to check the status, and confirm it is in one of the three states parsed, indexing, or indexed and is consistent with the interface display.
- Check file records and data rows separately, and confirm that the identifiers of files and data rows are separate and that rebuild tasks only cover data rows whose indexing has been completed.