What this category's data looks like
This page addresses the compulsory education dropout prevention and control scenario, and the materials come from three types of ledgers: dropout risk ledgers, persuasion-to-return records, and home visit records. The ledgers are updated weekly, while persuasion-to-return and home visit records are updated per instance, and the fields include student name, student registration number, date, number of persuasion attempts, signature, and receipt. When these three types of materials enter this function, they play different roles: the dropout risk ledger provides the risk list and division of responsibilities, the persuasion-to-return records provide the handling process and results, and the home visit records provide family reasons and follow-up visit information. During exception tracing, the name, quantity, and date of the same student in the three materials must be compared item by item to find inconsistencies; during manual review, the reviewer signs to confirm based on the original records, forming an exception list and review conclusion, which are handed to the responsible person for handling. If material versions are not synchronized, or signatures or receipts are missing, the tracing chain will be interrupted, and it is necessary to return to the source materials for correction.
Configuration and implementation checks
This part is executed by the memory turns and prompt configuration in the application settings, and outputs the referenced historical message range and answer constraints. The exception list and review conclusion are then completed by the model. The external entry point and authorization are handled by the publishing service and file permissions.
| Object | Check | Verification action |
|---|---|---|
| Memory turns | Match the range of prior context required by the question | Check the current value in the application settings, and run a trial with two consecutive follow-up questions to observe whether the second turn inherits the reference from the first turn |
| Dropout risk ledger | Whether the fields, dates, and signatures are complete, and whether the names, quantities, and dates are consistent with the persuasion-to-return records and home visit records | Extract one student record, check it item by item against the three materials, mark inconsistencies, and record the chatId |
| Persuasion-to-return records | Whether the number of persuasion attempts, dates, and signature receipts are consistent with the ledger | Search by student name and student registration number, check the number of attempts and dates, check whether receipts are missing, and record the dialogue trace in messages |
| Home visit records | Whether the home visit dates, reasons, and signatures are consistent with the ledger and persuasion-to-return records | Compare the home visit dates with the ledger dates, check the signatures, and if necessary use variables to mark items pending review |
Three common errors
- Phenomenon: the name or quantity of the same student does not match between the ledger and the persuasion-to-return records. Cause: the two materials were entered at different times, and no synchronization check was performed.
- Phenomenon: the home visit record date is inconsistent with the ledger date, and the tracing chain is broken. Cause: the versions were not synchronized, and the old material was not replaced.
- Phenomenon: the persuasion-to-return records lack signatures or receipts, and the review cannot be confirmed. Cause: signatures were not added in time during the handling stage, and material archiving was incomplete.
How to verify the setup
This function's acceptance: ask two consecutive follow-up questions, see whether the second question inherits the reference from the first question, and check whether chatId, messages, and variables are correctly referenced.
Business test: create one exception record, see whether it can be traced back to the source materials, and confirm that the exception list and review conclusion are supported by original records and signed by the reviewer.
- Check the current value of memory turns in the application settings, and run a trial with two consecutive follow-up questions to confirm that the second turn inherits the reference from the first turn.
- Extract one student record, check the name, quantity, date, and signature item by item against the dropout risk ledger, persuasion-to-return records, and home visit records, and mark inconsistencies.