What this category's data looks like
This page addresses the compulsory education dropout prevention and control scenario; materials mainly come from three types: dropout risk ledgers, persuasion-to-return records, and home visit records. Ledger fields include student identifier, risk level, determination date, and handler signature; persuasion-to-return records include number of persuasion attempts, method, result, and receipt; home visit records include visit date, participants, problem description, and parent signature. These materials are updated by semester or by month, and versions are synchronized by school or township. When entering the Q&A intake and missing-item follow-up function, the ledger plays the role of the base record for risk subjects and responsibility attribution, the persuasion-to-return records play the role of process and result evidence, and the home visit records play the role of family reasons and communication proof. The three types of materials corroborate one another; the same student's name, quantity, date, and signature must be consistent. If versions are not synchronized or a receipt is missing, the intake record and the completed follow-up items will first ask follow-up questions before answering, and then be transferred to the handler.
Configuration and implementation checks
This link is executed by the memory turns and prompt configuration in the application settings, and outputs the referenced historical message range and answer constraints. The "intake record and completed follow-up items" are then completed by the model. Memory turns control how many previous dialogue turns can be referenced at most when answering; if the value is too small, the preceding text cannot be used, and if the value is too large, irrelevant content is brought in and consumption increases.
| Object | Check | Verification action |
|---|---|---|
| Dropout risk ledger | Name, quantity, date, and signature are consistent across the three types of materials | Retrieve the ledger, persuasion-to-return records, and home visit records and compare item by item, checking the student identifier, risk level, determination date, and handler signature, and confirm that chatId is associated with the same session |
| Persuasion-to-return records | Number of persuasion attempts, date, and receipt correspond to the ledger | Check the number of persuasion attempts, method, result, receipt, and ledger date, verify whether the name and quantity of the same record fail to match across the two materials, and confirm whether the historical message range in messages covers the preceding text |
| Home visit records | Visit date, participants, and parent signature are complete | Check the visit date, participants, problem description, and parent signature, inspect cases where versions are not synchronized or signatures are missing, and confirm whether variables carry the referent from the preceding text |
| Intake record and completed follow-up items | Missing items are identified and followed up, and answers are based on materials | Run a trial with two consecutive follow-up questions, see whether the second question inherits the reference from the first question, and check whether the follow-up items correspond to the ledger, persuasion-to-return, and home visit fields |
Three common errors
- Phenomenon: the name, quantity, and date of the same record fail to match between the ledger and the persuasion-to-return records. Cause: the two places were entered separately and were not checked against the same student identifier and date, and versions were not synchronized.
- Phenomenon: home visit records lack a parent signature or receipt, but are treated as complete materials at intake. Cause: the signature and receipt fields were not validated during material collection, and the follow-up rules did not cover this missing item.
- Phenomenon: when the second question uses a reference, the answer does not address the question, responses become slower, and consumption rises. Cause: too many memory turns bring in irrelevant preceding text, or the opening message is treated as preceding text, and the historical message range does not match the preceding text range required by the question.
How to verify the setup
Acceptance for this function: run a trial with two consecutive follow-up questions, see whether the second question inherits the reference from the first question, and check whether chatId, messages, and variables output the referenced historical message range and answer constraints according to the memory turns and prompt configuration in the application settings.
Business test: ask a question missing key information, see whether it first follows up before answering, and whether the follow-up items fall on the fields, dates, and signatures of the ledger, persuasion-to-return records, and home visit records.
- In the application settings, check the current value of memory turns, match it with the preceding text range required by the question, and run a trial with two consecutive questions to test continuous follow-up.
- Compare the usage records of two runs, confirm that run consumption changes accordingly after adjusting the number of turns, and check whether the opening message is treated as preceding text.