```json { "classifications": [ { "column": "id", "role": "identifier", "confidence": 1.0, "reason": "Primary key with unique values per row (444 distinct in 444 rows)" }, { "column": "tenant_id", "role": "dimension", "confidence": 0.95, "reason": "Low cardinality (1 value), used for filtering/grouping by tenant" }, { "column": "dataset_id", "role": "identifier", "confidence": 1.0, "reason": "Foreign key reference to dataset, used for JOINs" }, { "column": "batch_id", "role": "identifier", "confidence": 1.0, "reason": "Foreign key reference to batch, used for JOINs" }, { "column": "engine", "role": "dimension", "confidence": 1.0, "reason": "Low cardinality categorical (3 values: lucene, tantivy, whoosh), used for grouping/filtering" }, { "column": "index_name", "role": "dimension", "confidence": 0.9, "reason": "Medium cardinality (97 distinct), categorical identifier used for grouping/filtering" }, { "column": "ok", "role": "dimension", "confidence": 1.0, "reason": "Boolean flag (0/1) indicating success/failure, used for filtering and grouping" }, { "column": "docs_indexed", "role": "measure", "confidence": 1.0, "reason": "Numeric quantity to aggregate (count of documents), suitable for sum/avg" }, { "column": "build_ms", "role": "measure", "confidence": 1.0, "reason": "Numeric duration metric to aggregate (build time in milliseconds), suitable for sum/avg" }, { "column": "index_bytes", "role": "measure", "confidence": 1.0, "reason": "Numeric size metric to aggregate (index size in bytes), suitable for sum/avg" }, { "column": "error_text", "role": "attribute", "confidence": 1.0, "reason": "Free-form descriptive text (error message), displayed but not aggregated" }, { "column": "created_at", "role": "dimension", "confidence": 1.0, "reason": "Timestamp for filtering, grouping by time periods, and temporal analysis" } ] } ```