Yuanbao College Entrance Exam Regional Passing Score Query

Problem

Regional college entrance exam score queries often contain ambiguous conditions. A user may ask for “Shanghai last year arts undergraduate” or “Fujian junior college,” but the backend expects normalized province, year, subject track, and admission batch. Direct keyword matching can miss aliases, city-to-province mapping, typo tolerance, subject-mode conversion, and batch expansion.

How It Works

The skill extracts four fields from the query: place, year, student, and luqupici.
- Place: matches the standard region against ./ref/place.json, supporting aliases such as “Fuzhou → Fujian” and “Gui → Guangxi.”
- Year: resolves explicit years or relative terms such as “this year” and “last year” into four-digit values; if absent, the field is omitted.
- Subject: uses the region and year to determine whether the province uses traditional liberal arts/sciences, 3+3, or 3+1+2, then maps terms such as “liberal arts” or “science” to the standard subject.
- Batch: matches admission batches against ./ref/luqupici_summary.json; if the user only says “undergraduate” or “junior college,” it expands to the relevant batch categories, and multiple batches are separated by spaces.
After extraction, it calls ./fetch_data.py to retrieve real data, keeps records consistent with the requested batch, and returns a concise answer.

Limits And Notes

This skill is useful for region-level passing score queries and query normalization. It may not cover every institution, major group, or enrollment plan detail. If key conditions such as region, subject, or batch are missing, the skill may ask a follow-up question or return a broader result set. Subject naming varies by province and year, so interpretations should be read together with the region, year, and batch.