What this category's data looks like
Modern apprenticeship and order class management involves three types of materials: school-enterprise cooperation agreements, order class training plans, and master-apprentice mentoring records. School-enterprise cooperation agreements contain the cooperation period, number of positions, and signature fields; order class training plans contain course modules, semester arrangements, and enterprise mentor confirmation; master-apprentice mentoring records contain mentoring dates, process names, and signatures of the master and apprentice. In terminology usage, these three types of materials often show multiple ways of writing the same concept, such as the mixed use of "post internship" and "job internship," making it difficult for translated names to refer back to the source text. When the materials enter the knowledge base retrieval function, agreements and plans serve as the main sources for semantic recall and full-text recall, while mentoring records serve as supplementary fragments, each playing the role of supplying candidate fragments. The update rhythm rolls forward according to semesters and agreement cycles, and the field characteristics focus on dates, quantities, and signatures for verification.
Configuration and implementation checks
This step is executed by the knowledge base retrieval service (retrieval node and search test), outputting candidate fragments and similarity scores (after mode, merge, rerank, filter, and upper-limit trimming). The "unified terminology glossary and comparison draft" is then completed by subsequent nodes or manually.
| Object | Check | Verification action |
|---|---|---|
| School-enterprise cooperation agreement | Whether the cooperation period, number of positions, and signature fields are consistent with the order class training plan | In the search test, test once each with synonymous questions and check whether the candidate fragments cover the agreement and plan; check the changes in the number of candidates and scores after switching searchMode |
| Order class training plan | Whether the course modules and semester arrangements match the dates and process names in the master-apprentice mentoring records | Compare the ranking of the top few items before and after enabling usingReRank; record the number of candidates and scores, and check whether the quoted text retains the key paragraphs |
| Master-apprentice mentoring records | Whether the mentoring dates and signatures of the master and apprentice are consistent with the names, quantities, and dates in the agreement and plan | Retrieve the same term using different expressions and see whether the candidates are stably hit; adjust limit to observe changes in the number of quoted items, and record the array length separately |
| Unified terminology glossary and comparison draft | Whether the same term is written consistently throughout the whole text, and whether the translated name can refer back to the source text | Sample three terms and check their writing throughout the whole text; use similarity to filter low-scoring candidates, and check the tokens of the quoted text and the number of array items and record them separately |
Three common errors
- Phenomenon: The names, quantities, and dates of the same record do not match between the agreement and the plan. Cause: The materials were drafted separately, without terminology unification and field alignment.
- Phenomenon: The version of the order class training plan is not synchronized, and the old course modules are still recalled. Cause: The update rhythm is not linked to the agreement cycle, and the knowledge base is not replaced in time.
- Phenomenon: The master-apprentice mentoring records lack signatures or receipts, and the quoted fragments cannot refer back to the source text. Cause: Fields are missing in the scanned copies, and retrieval only returns candidate fragments, not the complete document.
How to verify the setup
This function acceptance: test once each with synonymously rewritten questions, compare candidates and scores, then toggle rerank to see the ranking of the top few items; switch searchMode to observe differences in recall channels, and adjust limit to record the number of quoted items.
Business test: sample three terms and check whether their writing is consistent throughout the whole text, and check whether the translated names can refer back to the source text.
- In the knowledge base search test, test once each with synonymous questions, record the number of candidates and scores, and compare the order of the top few items before and after enabling rerank.
- Sample three terms and check their writing throughout the whole text, and verify whether the names, quantities, dates, and signatures in the agreement, plan, and mentoring records are consistent.