What this category's data looks like
Vocational school student internship management involves three types of materials: internship agreements, internship plans, and internship logs. Internship agreements contain the unit and school names, position, term, insurance, and signatures and seals; internship plans contain the batch, rotation arrangements, supervising teachers, and dates; internship logs contain the daily position, tasks, duration, mentor signature, and receipt. The three types of materials are filled in by different roles, with differing update rhythms: agreements are signed per batch, plans are adjusted per semester, and logs are recorded daily. When the materials enter the document parsing and chunking function, agreements and plans serve as basis sources, while logs serve as process evidence; each is split into retrievable data rows and bound to a unique data identifier, providing the foundation for subsequent basis entries to refer back to the original text.
Configuration and implementation checks
This step is executed by the parsing and training task (server-side task), outputting data rows and index status (parsed, indexing, indexed). "Basis entries that can refer back to the original text" are then completed by retrieval and subsequent processing.
| Object | Check | Verification action |
|---|---|---|
| Internship agreement | Fields must include unit name, position, term, insurance, signatures of both parties, and date, and must be consistent with the corresponding batch in the internship plan | Check the unit name, batch date, signatures, and receipts in the agreement and the plan separately, and confirm that the same record matches across the two materials |
| Internship plan | Fields must include batch, rotation position, supervising teacher, start and end dates, and signature, and the version must be the current one | Check the plan version against the agreement date, confirm that no old version is mixed in, and check whether the signature fields are complete |
| Internship log | Fields must include daily position, tasks, duration, mentor signature, and receipt, and the dates must be continuous and match the plan period | Check the log dates against the plan period one by one, and confirm that records missing signatures or receipts have been marked |
| Data rows and index status | Is in one of the three states parsed, indexing, or indexed and is consistent with the interface display; data pending indexing does not participate in recall | Open the data rows to check the status, and use chunkSize, chunkSplitter, and trainingType to check chunking, splitting, and training mode respectively |
Three common errors
- Symptom: The position name and duration of the same internship record do not match between the agreement and the log. Cause: The two materials are filled in by different roles, and dates and quantities were not cross-checked.
- Symptom: An old version of the internship plan is still cited as the business basis. Cause: The version was not synchronized, the current version and historical versions were not distinguished, and old data was mixed in during retrieval.
- Symptom: The log lacks a mentor signature or receipt, and the basis entry cannot refer back to the original text. Cause: Material archiving is incomplete; although the parsed data rows are visible, they lack a verifiable signature field.
How to verify the setup
This feature's acceptance: 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 the rebuild covers only data whose indexing has been completed.
Business test: ask a basis question that spans materials, and see whether the answer provides a traceable source.
- Open the data rows to check the status, confirm consistency with the interface display, and record the number of array entries and the tokens of the referenced text; record the number of citations and the citation content budget separately.
- Check the file records and data rows separately, confirm that the identifiers of files and data rows are distinguishable, and check whether the data rows covered by the rebuild task include only data whose indexing has been completed.