What this category's data looks like
This page addresses the compulsory education dropout prevention and school retention scenario, with materials drawn from three types: dropout risk ledgers, persuasion-to-return records, and home visit records. The ledger provides students' basic information and risk levels, with fields including name, student registration number, risk category, determination date, and responsible person; the persuasion-to-return records carry the number of persuasion attempts, persuasion methods, parent signatures, and receipts; the home visit records supplement the household visit time, visiting teacher, parent opinions, and signatures. The update cadence is weekly or event-triggered, with persuasion and home visits recorded as they occur. When these three types of materials enter the knowledge base retrieval node, each plays a different role: the ledger is used to compare the risk list and versions, the persuasion-to-return records are used to verify the number of attempts, dates, and signatures, and the home visit records are used to trace back to the household visit facts and receipts. Field characteristics are concentrated in four categories: dates, names, quantities, and signatures, which must be aligned item by item during handover.
Configuration and implementation checks
This step is executed by the citation output of the knowledge base retrieval node, which outputs an array of cited content, each entry containing the original text snippet and source identifier, and the length may be 0. The "handover-ready deployment status and operation records" are completed subsequently by AI nodes, code nodes, or manual work.
| Object | Check | Verification action |
|---|---|---|
| Dropout risk ledger | Fields include name, student registration number, risk category, determination date, and responsible person, and are consistent with the names, quantities, and dates in the persuasion-to-return records and home visit records | In the node, check the selected knowledge base collection and vector model, and compare item by item the name, date, and signature entries in the ledger against the persuasion-to-return and home visit records |
| Persuasion-to-return records | The number of persuasion attempts, persuasion methods, parent signatures, and receipts are complete, and match the risk list in the ledger and the dates in the home visit records | Open the search parameter panel and check the values item by item, and verify whether the cited snippets can trace back to the original persuasion-to-return paragraphs and signature positions |
| Home visit records | The household visit time, visiting teacher, parent opinions, and signatures are complete, and do not conflict with the determination date in the ledger or the number of attempts in the persuasion-to-return records | Run a test with a question that has no matching materials, see whether the output is empty while the pipeline continues, and record the number of entries in the citation array and the quoteMaxToken budget |
Three common errors
- Symptom: The name or date of the same student does not match between the ledger and the persuasion-to-return records. Cause: The two materials were entered by different shifts, were not backfilled according to the same determination date and student registration number, and item-by-item verification was missing during handover.
- Symptom: The ledger version was not synchronized, and the old risk list is still being cited. Cause: After the knowledge base was updated, the selected collection and vector model were not verified, and the handover form did not record the version change and effective time.
- Symptom: The persuasion-to-return records lack a parent signature or receipt, and the home visit records have no signature. Cause: When archiving materials, only the body text was entered without adding the signature page, and the signature items were not checked one by one during handover.
How to verify the setup
This feature's acceptance: ask a question with original text evidence, and check whether the citation array and footnote markers can trace back to the original paragraphs; then ask a question with no matching materials, and confirm that the array is empty while the pipeline runs through.
Business testing: verify item by item once according to the handover form, see which record is missing, and focus on checking whether the version, connection, and task records are consistent.
- Open the search parameter panel and check the values item by item, confirm they are consistent with the settings in the knowledge base retrieval section, and record the number of entries in the citation array and the
quoteMaxTokenbudget. - Open the citation footnote markers in the conversation, check whether the cited snippets correspond to the original text against the source documents, and also inspect the source identifiers in
responseData(non-streaming module results returned with detailed output enabled) andquoteList.