What this category's data looks like
Vocational college program setup filing materials mainly consist of program setup filing forms, talent demand justification, and faculty and equipment lists. They come from the academic affairs and human resources departments of the institutions and are updated by academic year or application batch. The filing form contains the program name, program length, planned enrollment number, and signature date; the justification materials contain descriptions of industry job demand, survey dates, and expert signatures; the faculty and equipment list contains the number of teachers, title structure, equipment units, and procurement dates. After these three types of materials enter the document parsing and chunking process, each plays a different role: the filing form serves as the anchor for the main record, the justification materials provide demand-side evidence, and the faculty and equipment list provides support for institutional conditions. Parsing and chunking are split according to chunkSize and chunkSplitter, trainingType determines the training mode, and after parsing is completed the materials are immediately stored in the database as visible data rows for subsequent retrieval and discrepancy localization.
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 "discrepancy list" is then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Program setup filing form | The program name, planned enrollment number, and signature date are consistent across the three materials | Check the program name, quantity, and date in the filing form against the justification materials and the faculty and equipment list respectively, and record inconsistent fields; check whether the data row status is parsed or indexed |
| Talent demand justification | The survey date and the number of job demands correspond to the application batch of the filing form | Check whether the survey date in the justification materials and the signature date of the filing form are from the same batch, and check whether expert signatures and receipts are complete; check whether the fields after splitting by chunkSize and chunkSplitter are complete |
| Faculty and equipment list | The number of teachers and equipment units match the institutional conditions filled in the filing form | Check the number of teachers and equipment units in the list, check the procurement date and signature column, and confirm version synchronization; check whether the training mode corresponding to trainingType matches the material type |
| Data row status | It is in one of the three states parsed / indexing / indexed and is consistent with the interface display | Open the data row to check the status, confirm that data pending indexing does not participate in recall, and that rebuilding only covers indexed data |
Three common errors
- Symptom: The name, quantity, and date of the same program do not match across two materials. Cause: The filing form and justification materials are filled in separately, no field-level comparison is performed, and discrepancies are not exposed before review.
- Symptom: Material versions are not synchronized, and an old faculty and equipment list still participates in retrieval. Cause: After updating, parsing and database storage were not performed again, the data row status remains at the old index, and retrieval hits outdated content.
- Symptom: Filing materials lack signatures or receipts, and responsibility cannot be confirmed during review. Cause: Parsing and chunking only process the main text, and signature pages and receipts are not included in the field verification scope.
How to verify the setup
Upload one material, observe the data row status transitions and the coverage scope of the rebuild task, and confirm that the status transitions among parsed, indexing, and indexed, and that rebuilding only covers indexed data.
Use one real record to find fields that are inconsistent with the ledger.
- Check whether the data row status is parsed / indexing / indexed, and confirm that data pending indexing is not recalled.
- Check whether the data rows covered by the rebuild task include only data whose indexing has been completed.