What this category's data looks like
This category's data looks like this: The grade entry and credit recognition scenario involves three types of materials: transcripts, deferred examination applications, and make-up examination arrangements. Transcripts are exported from the academic affairs system, with fields including course name, credits, grade, grade point, and instructor signature; deferred examination applications are submitted by students, with fields including course, reason for deferral, date, and counselor signature; make-up examination arrangements are published by the college, with fields including make-up examination course, time, location, and invigilator signature. The material update cadence advances with the semester: transcripts are fixed after entry, deferred examination applications are archived after approval, and make-up examination arrangements are published before the examination. When the three types of materials enter the knowledge base, each serves as an independent data row, and an index is generated through the vector model to support subsequent retrieval and citation. The name, quantity, and date of the same record in the three materials must remain consistent, and must be updated synchronously when the version changes.
Configuration and implementation checks
Configuration and implementation checks: This step is executed by the index service and the vector store, 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 citation trimming only take supporting materials).
| Object | Check | Verification action |
|---|---|---|
| Transcript | Course name, credits, grade, grade point, and instructor signature are complete, and are consistent with the same course name and credits in the deferred examination application and make-up examination arrangement | Open the data row and check the index fields to confirm that multiple vector identifiers are complete; search the course name to see whether data in the same group is merged |
| Deferred examination application | Course, reason for deferral, date, and counselor signature are consistent with the corresponding course and date in the transcript, with no missing signature or receipt | In the knowledge base settings, check that the index model is consistent with the vector model actually used by the knowledge base; search the deferred examination course to see whether it can be recalled |
| Make-up examination arrangement | Make-up examination course, time, location, and invigilator signature match the same course name, quantity, and date in the transcript and deferred examination application | Check the rebuild task records to confirm that after changing the model, only data that has already been indexed is triggered; search the make-up examination course to see whether the same group is merged |
| Data row index fields | Multiple vector identifiers for one piece of data are all present, and the status is indexed | Open a data row and check the index fields to confirm that the number of vector identifiers corresponds to the data row; search once to see whether data in the same group is merged |
Three common errors
- Symptom: duplicate entries appear in retrieval results. Inspect the following: the course name, quantity, and date of the same record in the transcript and make-up examination arrangement do not match, and the vector identifiers are not merged.
- Symptom: recall quality drops after changing the model. Inspect the following: the rebuild task records do not cover all data that has already been indexed, the version is not synchronized, and old-version indexes remain.
- Symptom: a data row has content but cannot be retrieved. Inspect the following: the deferred examination application lacks a counselor signature or receipt, the index fields are missing, and vectors are not generated.
How to verify the setup
How to verify the setup: Acceptance for this feature: sample a data row and check the index fields to confirm that multiple vector identifiers are all present and the status is indexed; then search once to see whether data in the same group is merged and whether it can be recalled.
Business test: change one piece of content and see whether the version number and citations are updated synchronously; check the tokens and array count of the cited text, record the citation count separately, and check limit as needed.
- In the knowledge base settings, check that the index model is consistent with the vector model actually used by the knowledge base, and view the data scope covered by the rebuild task records.
- Open a data row and check the index fields to confirm that multiple vector identifiers are complete; search once to see whether data in the same group is merged and whether the status is indexed.