What this category's data looks like
This page addresses the course scheduling and lesson-hour deduction scenarios of training institutions. The materials include three types: class schedules, lesson-hour deduction records, and schedule-change requests. Class schedules are exported weekly from the academic affairs system, lesson-hour deduction records are submitted by the teaching instructor after class, and schedule-change requests are initiated by academic affairs staff or parents and retained after approval. The update rhythm is that class schedules are refreshed at a fixed weekly cadence, lesson-hour deduction records are appended in real time with each class session, and schedule-change requests are archived at approval nodes. Field characteristics center on course name, class, teacher, date and time slot, number of lesson hours, and signature or receipt status. When the materials enter the vector model and indexing feature, class schedules and lesson-hour deduction records are split as the primary basis text and generate vectors, while schedule-change requests are synchronized into the index as supplementary basis; together the three support cross-material basis queries and back-referencing.
Configuration and implementation checks
This step is executed by the indexing service and the vector store, outputting vectors and index status (whether recall is possible, whether data in the same group is merged). Basis entries that can back-reference the original text are then completed by subsequent AI nodes, code nodes, or manual work (retrieval and citation trimming take only supporting materials). Index status is based on indexed and rebuild task records, and rebuild task records cover the data range that has completed indexing.
| Object | Check | Verification action |
|---|---|---|
| Class schedule | Course name, class, teacher, date and time slot, and signature column are complete; the name and date of the same class session match across the two materials | Open one class schedule data row, verify whether multiple vector identifiers in the index field are all present, and check whether the same group is merged during retrieval |
| Lesson-hour deduction record | Number of lesson hours, deduction date, and teacher signature or receipt status are consistent with the class schedule, and the version is current | In the knowledge base settings, verify that the indexing model matches the vector model actually used by the knowledge base, and check the coverage of the rebuild task records |
| Schedule-change request | The dates before and after the change, the number of lesson hours, and the approval signature can back-reference the original text; entries missing a signature or receipt are marked separately | Retrieve one schedule-change request, confirm that after recall it is merged back into the original data by identifier, and that the highest score is taken for data in the same group |
| Data row index field | Multiple vector identifiers of one data entry are all present, the status is indexed, and the rebuild task record is queryable | Open one data row to verify the index field, and verify the budget records of limit in the cited text |
Three common errors
- Symptom: duplicate entries appear in retrieval results. Inspect the following: the name, quantity, or date of the same record does not match between the class schedule and the lesson-hour deduction record, so it is not recognized as the same group when vectors are merged back into the original data.
- Symptom: recall quality drops after changing models. Inspect the following: the version is not synchronized, and data rows whose indexing is not complete do not participate in recall; indexing must be completed first and then rebuilt.
- Symptom: a data row has content but cannot be retrieved. Inspect the following: the schedule-change request is missing a signature or receipt, vector identifiers in the index field are missing, and the status has not reached indexed.
How to verify the setup
Feature acceptance: sample one data row to inspect the index field, confirm that multiple vector identifiers are all present and the status is indexed, then retrieve once to see whether data in the same group is merged and whether the highest score is taken.
Business test: ask a lesson-hour quantity basis question that spans the class schedule and the lesson-hour deduction record, see whether the answer provides a source that can back-reference the original text, and verify that the number of citations and quoteMaxToken are recorded separately.
- In the knowledge base settings, verify that the indexing model matches the vector model actually used by the knowledge base, and check the data range covered by the rebuild task records.
- Open one data row to verify the index field, confirm that multiple vector identifiers are all present and the status is indexed, and separately record the number of citations and the tokens of the cited text.