What this category's data looks like
Materials for modern apprenticeship and order class management usually fall into three categories: school-enterprise cooperation agreements, order class training plans, and master-apprentice records. School-enterprise cooperation agreements contain the cooperation period, positions, number of people, and signature pages of both parties; order class training plans contain course modules, class hours, assessment methods, and approval dates; master-apprentice records contain mentoring masters, apprentice lists, week numbers, work hours, and receipt signatures. Materials are mostly updated by semester or training cycle, and version numbers, revision dates, and signature fields are field characteristics. After entering this function, school-enterprise cooperation agreements and order class training plans are parsed and chunked as basis sources to answer questions about positions, number of people, courses, and assessment criteria; master-apprentice records are parsed and chunked as process materials to verify the facts of mentoring and the completeness of receipts. The three types of materials are each stored as data rows for subsequent retrieval and citation.
Configuration and implementation checks
This step is executed by parsing and training tasks (server-side tasks), outputting data rows and index status (parsed, indexing, indexed). "Basis entries that can be traced back to the original text" are then completed by retrieval and subsequent processing. After parsing is completed, parsed chunks are immediately stored as visible data rows and bound to a unique data identifier; the index transitions through three states, and data pending indexing does not participate in recall.
| Object | Check | Verification action |
|---|---|---|
| School-enterprise cooperation agreement | The cooperation period, positions, number of people, and signature pages match the original text; the version is the current version; fields and dates do not conflict with other materials | Open the data row and check whether the status is parsed, indexing, or indexed; check the signature fields and receipts; check whether the chunks produced by chunkSize and chunkSplitter cover the position and number-of-people fields |
| Order class training plan | Course modules, class hours, assessment methods, and approval dates can be traced back to the original text; the version number and revision date are current; names and quantities do not mismatch with other materials | Check the file record and the data row separately; check the approval date and signature; check whether the processing mode corresponding to trainingType covers this material |
| Master-apprentice record | Mentoring masters, apprentice lists, week numbers, work hours, and receipt signatures match the original text; entries missing signatures or receipts are marked separately | Open the data row and check the status; check the receipt signature and date; check whether chunking keeps week numbers and work hours in the same data row |
Three common errors
- Phenomenon: The name, quantity, and date of the same record do not match between the school-enterprise cooperation agreement and the order class training plan. Cause: The two materials were revised separately, field definitions were not unified, and after parsing and chunking they became separate rows without cross-checking.
- Phenomenon: The training plan has been revised to a new version, but retrieval still returns the old course and class hours. Cause: The old data row is still in the completed indexing state; rebuilding only targets data that has completed indexing, and the version was not synchronized.
- Phenomenon: Master-apprentice records missing signatures or receipts are still cited as basis entries. Cause: Parsing and chunking only split by content, and signature and receipt fields were not checked as separate data rows.
How to verify the setup
Acceptance for this function: upload one material, observe the data row status transitions and the coverage scope of the rebuild task; check whether the status is parsed, indexing, or indexed, check that the rebuild task only covers data rows that have completed indexing, and check the chunking and processing modes corresponding to chunkSize, chunkSplitter, and trainingType.
Business test: ask a cross-material basis question, see whether the answer provides traceable sources, and check the number of citations and the tokens of the cited text; record the array length separately.
- Open the data row and check the status, confirm it is in one of the three states and consistent with the interface display; check the identity identifiers of the file record and the data row separately.
- View the data rows covered by the rebuild task, confirm that only data that has completed indexing is covered; separately mark master-apprentice records missing signatures or receipts and trace them back to the original text.