What this category's data looks like
This page addresses the deployment, operations, and application handover stage in the scenario of credit management for primary and secondary school teacher training, involving three types of materials: training notices, credit recognition forms, and participant lists. A training notice determines the training project, training hours, and the criteria for credit recognition, and serves as the date and quantity baseline for the credit recognition form and the participant list; the credit recognition form carries teacher names, training names, recognized credits, and a signature field, and is key to verifying versions and receipts during handover; the participant list records participating teachers, their schools, and attendance status, and must match the recognition form item by item in number of people, names, and dates. The materials enter the knowledge base retrieval and recall function in the form of files or tables, becoming the source of candidate chunks; the update cadence follows training batches and recognition cycles, and the field characteristics center on project name, teacher name, credit amount, recognition date, and signature receipt.
Configuration and implementation checks
This step is executed by the knowledge base retrieval service (retrieval node and search test), outputting candidate chunks and similarity scores (after mode, merge, re-rank, filter, and upper-limit trimming). The handover-ready deployment state and operation records are completed by subsequent nodes or manually.
| Object | Check | Verification action |
|---|---|---|
| Training notice | The project name, training date, training hours, and credit recognition criteria must correspond item by item with the credit recognition form | In the search test, test once each with the project name and the date as questions, check whether the project name, date, and credit criteria in the candidate chunks are consistent, and record the number of candidates and the score |
| Credit recognition form | Teacher names, training names, recognized credits, recognition dates, and signature receipts must be consistent with the participant list | Check names, quantities, dates, and signatures item by item according to the handover sheet, switch searchMode to compare candidate differences among the three options of semantic, full-text, and hybrid, and record the number of candidates and the score |
| Participant list | Participating teachers, their schools, attendance status, and number of people must correspond with the recognition form | Retrieve with teacher names and schools as questions, check the number of people and attendance status in the candidate chunks, turn usingReRank on and off to compare the ranking of the first few results, and record the number of candidates and the score |
| Deployment state and operation records | Version, connection, and task records are consistent, and the retrieval mode and re-rank switch records are traceable | Check version, connection, and task records item by item according to the handover sheet, inspect the recorded values of similarity and limit, and confirm that referenced text tokens and array counts are registered separately |
Three common errors
- Symptom: The name or number of people for the same teacher does not match between the recognition form and the participant list. Cause: The versions of the two materials were not synchronized, or they were not aligned by the same project name and date during data entry.
- Symptom: The handover sheet lacks a signature or receipt, and operations cannot confirm the deployment state. Cause: The signature field on the recognition form was not filled in, and the operation records were not handed over together with the version.
- Symptom: When the question is phrased differently, key chunks cannot be recalled, and although there are many candidates, the ones ranked first are irrelevant. Cause: The retrieval mode does not match the material form, re-rank is not enabled, or similarity filtering and referenced upper-limit trimming are too tight.
How to verify the setup
This feature's acceptance: test once each with synonym-rewritten questions, compare the number of candidates and the score, then turn re-rank on and off to observe the ranking of the first few results; switch searchMode to observe candidate differences among semantic, full-text, and hybrid, and after adjusting similarity and limit, record referenced text tokens and array counts.
Business test: check once item by item according to the handover sheet to see which record is missing; check whether the names, quantities, dates, and signature receipts of the training notice, credit recognition form, and participant list are consistent, and whether version, connection, and task records are synchronized.
- In the knowledge base search test, construct questions respectively from the project name and date of the training notice, the teacher names and credits of the recognition form, and the schools and attendance status of the participant list, test each once, and record the number of candidates and the score.
- Check version, connection, and task records item by item according to the handover sheet, confirm that the values of
searchMode,usingReRank,similarity, andlimithave been registered, record referenced text tokens and array counts separately, and after filling in missing items, proceed with the handover.