What this category's data looks like
This page focuses on data versions and change maintenance in the course selection, scheduling, and classroom resource allocation scenario. Material sources include three categories: course offering plans, scheduling tables, and course selection rules. The update cadence is centralized release before the semester, with rolling revisions during the semester based on schedule changes and supplementary selection. Course offering plans carry course names, credits, capacities, and instructor signatures; scheduling tables carry classrooms, time slots, week numbers, and academic affairs receipts; course selection rules carry selection targets, priorities, and effective dates. When these three categories of materials enter the vector model and index feature, each is written as an independent data row, and the index field records the corresponding multiple vector identifiers. Course offering plans focus on course attributes, scheduling tables focus on spatiotemporal occupancy, and course selection rules focus on constraint conditions. Together, the three support traceability and reference synchronization of the same record in the version ledger.
Configuration and implementation checks
This step is executed by the index service and vector database, outputting vectors and index status (whether recall is possible, whether data in the same group is merged). The version ledger and change records are then completed by subsequent AI nodes, code nodes, or manual work (retrieval and reference trimming only take supporting materials). The index service is only responsible for mapping the vectors produced after material splitting to the original data identifiers; it does not generate glossaries, comparison drafts, or version ledgers.
| Object | Check | Verification action |
|---|---|---|
| Course offering plan | Course name, credits, capacity, and instructor signature are unique in the current version, and the old version is traceable | Open a course offering plan data row and check whether the index field covers multiple vector identifiers and whether the status is indexed |
| Scheduling table | Classroom, time slot, week number, and academic affairs receipt are consistent in the current version, and references are synchronized | View the rebuild task record and confirm that after changing the model, rebuild is triggered only for data that has already completed indexing |
| Course selection rules | Selection targets, priorities, and effective dates are unique in the current version, and the old version is traceable | Check whether the index model is consistent with the vector model actually used by the knowledge base, then retrieve data in the same group to see whether it is merged |
| Version ledger and change records | The current version is unique, the old version is traceable, and references are synchronized; change one piece of content and see whether the version number and references are synchronized | Sample a data row to view the index field, then retrieve once to see whether the same group is merged |
Three common errors
- Symptom: duplicate entries appear in retrieval results. Inspect the following: the name, quantity, and date of the same record do not match between the course offering plan and the scheduling table, and the vectors hit different original data respectively.
- Symptom: recall quality drops after changing the model. Inspect the following: the version is not synchronized, rebuild is performed, and data rows that have not completed indexing do not participate in recall; indexing must be completed first, then rebuild.
- Symptom: a data row has content but cannot be retrieved. Inspect the following: the course selection rules lack a signature or receipt, the material has not entered the current version, and the index field does not record the corresponding vector identifier.
How to verify the setup
This feature acceptance: sample a data row to see whether the index field contains multiple vector identifiers, then retrieve once to see whether the same group is merged; check the status value indexed and the rebuild task record, the tokens of the reference text are recorded separately, the number of array entries is recorded separately, and limit are checked according to the knowledge base settings.
Business test: change one piece of content and see whether the version number and references are updated synchronously
- In the knowledge base settings, check whether the index model is consistent with the vector model actually used by the knowledge base, and confirm the values of
defaultTokenandmaxToken - Open a data row to check whether the index field records multiple vector identifiers, and view the rebuild task record to confirm the coverage scope