What this category's data looks like
This page addresses the international student admissions and enrollment management category. Materials come from admissions brochures, admission letters, and visa application forms, and the update cadence follows admissions batches and enrollment status changes. Field characteristics center on name spelling, date of birth, passport number, admitted program, and signature and acknowledgment fields. When entering this feature, admissions brochures and visa application forms provide the source for multilingual terminology and field mapping, while admission letters carry the final publication wording; the indexing service and vector database are responsible for splitting multilingual text into recallable units, with the original data stored in business database data rows and multiple vector identifiers recorded in the index field. Because materials play different roles, the same record may be scattered across different documents, so field and date consistency must be confirmed first before entering the vectorization and retrieval stage.
Configuration and implementation checks
This step is executed by the indexing service and vector database, outputting vectors and index status (whether recall is possible and whether data in the same group is merged). The "unified terminology table and mapping draft" is then completed by subsequent AI nodes, code nodes, or manual work (retrieval and citation trimming only take supporting materials). The indexing model must be consistent with the vector model actually used by the knowledge base, and rebuild tasks only cover data that has already been indexed.
| Object | Check | Verification action |
|---|---|---|
| Admissions brochure | Name spelling, date of birth, passport number, admitted program, and signature field must be consistent with the admission letter | Open the brochure and admission letter, compare field by field the name, quantity, and date of the same record, and check whether the signature or acknowledgment is complete |
| Admission letter | The name, passport number, program name, and date in the visa application form must be traceable back to the original text of the admission letter | Extract three terms, check whether their spelling is consistent throughout the document, and verify whether the versions of the admission letter and application form are synchronized |
| Visa application form | The indexing model is consistent with the vector model actually used by the knowledge base, and the data row index field contains all vector identifiers | Check the indexing model in the knowledge base settings, open a data row to check the index field, and confirm the status is indexed |
| Data row | After changing models, rebuild is triggered only for data that has already been indexed, and the data scope covered by the rebuild task record must be checked | Check the coverage scope of the rebuild task record, run a retrieval to confirm that data in the same group is merged, and check the defaultToken and maxToken budgets |
Three common errors
- Symptom: The name, quantity, or date of the same record does not match between the admissions brochure and the admission letter. Inspect the following: Multilingual materials are maintained separately, and fields have not been given a unified terminology table and mapping draft.
- Symptom: The versions of the visa application form and admission letter are not synchronized, and signatures or acknowledgments are missing. Inspect the following: Versions and signature fields were not checked by batch before publication, and acknowledgments were not archived.
- Symptom: Duplicate entries appear in retrieval results, and recall quality drops after changing models. Inspect the following: The data row index field does not record all vector identifiers, and the rebuild task record does not cover data that has already been indexed.
How to verify the setup
Acceptance for this feature: extract one data row and inspect the index field to confirm that multiple vector identifiers are all present and the status is indexed; then run a retrieval once to see whether data in the same group is merged and whether the vector score takes the highest score.
Business test: extract three terms and check whether their spelling is consistent throughout the document, and verify the name, quantity, date, and signature acknowledgment across the admissions brochure, admission letter, and visa application form.
- Check in the knowledge base settings that the indexing model is consistent with the vector model actually used by the knowledge base, and record the
defaultTokenandmaxTokenbudgets. - Open a data row to check the index field, inspect the coverage scope of the rebuild task record, and confirm that after changing models, rebuild is triggered only for data that has already been indexed.