zhaoqingliang/data-ge
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

demo数据

Browse Source
This commit is contained in:
zhaoawd
2025-11-14 00:58:00 +08:00
parent 7eb3c059a1
commit a72ca3593e
13 changed files with 2733 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
demo/水务/水务-gemini2.5-ge-result.json Normal file
View File

@ -0,0 +1 @@
{"role": "dimension", "time": {"range": null, "column": null, "has_gaps": null, "granularity": "unknown"}, "grain": ["service_point_id"], "table": "data-ge.water_meter_info", "columns": [{"name": "meter_subtype", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 9, "distinct_ratio": 0.03, "pk_candidate_score": 0.03, "metric_candidate_score": 0.0}, {"name": "installation_position", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 4, "distinct_ratio": 0.013333333333333334, "pk_candidate_score": 0.013333333333333334, "metric_candidate_score": 0.0}, {"name": "supply_office", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 11, "distinct_ratio": 0.03666666666666667, "pk_candidate_score": 0.03666666666666667, "metric_candidate_score": 0.0}, {"name": "meter_diameter", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 8, "distinct_ratio": 0.02666666666666667, "pk_candidate_score": 0.02666666666666667, "metric_candidate_score": 0.0}, {"name": "account_id", "dtype": "unknown", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "该列的统计指标（如空值率、唯一性）缺失，但根据命名规则推断为ID。", "enumish": null, "null_rate": null, "top_values": [], "semantic_type": "id", "distinct_count": null, "distinct_ratio": null, "pk_candidate_score": 0.9, "metric_candidate_score": 0.0}, {"name": "service_point_id", "dtype": "unknown", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "该列的统计指标（如空值率、唯一性）缺失，但根据命名规则推断为ID。", "enumish": null, "null_rate": null, "top_values": [], "semantic_type": "id", "distinct_count": null, "distinct_ratio": null, "pk_candidate_score": 0.95, "metric_candidate_score": 0.0}, {"name": "station", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 36, "distinct_ratio": 0.12, "pk_candidate_score": 0.12, "metric_candidate_score": 0.0}, {"name": "meter_type", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 5, "distinct_ratio": 0.016666666666666666, "pk_candidate_score": 0.016666666666666666, "metric_candidate_score": 0.0}, {"name": "district", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 13, "distinct_ratio": 0.043333333333333335, "pk_candidate_score": 0.043333333333333335, "metric_candidate_score": 0.0}, {"name": "meter_status", "dtype": "string", "stats": {"max": null, "min": null, "std": null, "mean": null, "skewness": null}, "comment": "该列只有一个唯一值 '有效'。", "enumish": true, "null_rate": 0.0, "top_values": [], "semantic_type": "dimension", "distinct_count": 1, "distinct_ratio": 0.0033333333333333335, "pk_candidate_score": 0.0033333333333333335, "metric_candidate_score": 0.0}], "quality": {"warning_hints": ["列 'meter_status' 只有一个唯一值 '有效'，可能为常量列。"], "failed_expectations": []}, "row_count": 300, "fk_candidates": [], "confidence_notes": ["表角色(role)被推断为 'dimension'，因为其列几乎完全由ID和类别属性构成，且缺少数值指标或时间序列列。", "主键候选(primary_key_candidates) 'service_point_id' 和 'account_id' 是基于命名约定（包含'_id'）推断的。其唯一性和非空性未在GE结果中直接度量，因此这是一个高置信度的猜测。", "表粒度(grain)可能为 'service_point'，与推断的主键 'service_point_id' 相对应。", "未根据列名或数据格式识别出时间列。"], "primary_key_candidates": [["service_point_id"], ["account_id"]]}

180
demo/水务/水务-gemini2.5-snippet-alias.json Normal file
View File

@ -0,0 +1,180 @@
[
{
"id": "snpt_count-service-points-by-dimension",
"aliases": [
{
"text": "各个区有多少水表",
"tone": "口语"
},
{
"text": "按维度统计用水点数",
"tone": "中性"
},
{
"text": "各维度用水点数量分布",
"tone": "专业"
}
],
"keywords": [
"用水点数",
"service_point_count",
"数量",
"统计",
"汇总",
"aggregate",
"维度",
"dimension",
"区域",
"district",
"供水所",
"分组统计",
"水表"
],
"intent_tags": [
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_topn-service-points-by-dimension",
"aliases": [
{
"text": "哪个地方水表最多",
"tone": "口语"
},
{
"text": "用水点数Top-N排名",
"tone": "中性"
},
{
"text": "Top-N用水点数维度排行",
"tone": "专业"
}
],
"keywords": [
"Top-N",
"top",
"排名",
"排行",
"ranking",
"最多",
"用水点数",
"service_point_count",
"维度",
"dimension",
"站点",
"station",
"水表"
],
"intent_tags": [
"topn",
"by_dimension"
]
},
{
"id": "snpt_ratio-service-points-by-dimension",
"aliases": [
{
"text": "各种水表各占多少",
"tone": "口语"
},
{
"text": "各维度用水点数占比",
"tone": "中性"
},
{
"text": "用水点维度构成分析",
"tone": "专业"
}
],
"keywords": [
"占比",
"percentage",
"百分比",
"ratio",
"构成",
"分布",
"用水点数",
"水表类型",
"meter_type",
"维度",
"dimension",
"水表"
],
"intent_tags": [
"ratio",
"by_dimension"
]
},
{
"id": "snpt_quality-check-duplicate-spid",
"aliases": [
{
"text": "有没有重复的水表号",
"tone": "口语"
},
{
"text": "检查重复的用水点ID",
"tone": "中性"
},
{
"text": "用水点ID唯一性校验",
"tone": "专业"
}
],
"keywords": [
"数据质量",
"quality",
"检查",
"校验",
"重复",
"duplicate",
"唯一性",
"uniqueness",
"用水点ID",
"service_point_id",
"异常检测",
"主键"
],
"intent_tags": [
"quality",
"by_dimension"
]
},
{
"id": "snpt_sample-filter-service-points-by-dims",
"aliases": [
{
"text": "给我看城区的机械表",
"tone": "口语"
},
{
"text": "按多维度筛选用水点",
"tone": "中性"
},
{
"text": "多维组合条件过滤用水点",
"tone": "专业"
}
],
"keywords": [
"筛选",
"过滤",
"filter",
"查询",
"明细",
"列表",
"sample",
"用水点",
"区域",
"district",
"水表类型",
"meter_type",
"条件查询"
],
"intent_tags": [
"sample",
"filter"
]
}
]

186
demo/水务/水务-gemini2.5-snippet.json Normal file
View File

@ -0,0 +1,186 @@
[
{
"id": "snpt_count-service-points-by-dimension",
"desc": "按指定维度（如区域、供水所）分组，统计各分类下的用水点数量。",
"type": "aggregate",
"title": "按维度统计用水点数",
"examples": [
"按区域统计用水点数量",
"各个供水所分别有多少个用水点"
],
"variables": [
{
"name": "dimension_column",
"type": "column",
"default": "district"
}
],
"dialect_sql": {
"mysql": "SELECT\n `${dimension_column}`,\n COUNT(DISTINCT service_point_id) AS service_point_count\nFROM\n `data-ge.water_meter_info`\nGROUP BY\n `${dimension_column}`\nORDER BY\n service_point_count DESC;"
},
"applicability": {
"constraints": {
"notes": [
"适用于对水表档案信息进行分类汇总统计。",
"可将变量 ${dimension_column} 替换为任一维度列，如 district, supply_office, station, meter_type 等。"
],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "用水点数：对 `service_point_id` 进行去重计数，代表一个独立的服务点（通常对应一个水表）。统计粒度为“指定维度”。"
},
{
"id": "snpt_topn-service-points-by-dimension",
"desc": "按指定维度（如区域、站点）统计用水点数，并展示数量最多的前N个分类。",
"type": "topn",
"title": "Top-N 用水点数维度排名",
"examples": [
"哪个区域的用水点最多",
"用水点数排名前5的站点是哪些"
],
"variables": [
{
"name": "dimension_column",
"type": "column",
"default": "station"
},
{
"name": "top_n",
"type": "int",
"default": 10
}
],
"dialect_sql": {
"mysql": "SELECT\n `${dimension_column}`,\n COUNT(DISTINCT service_point_id) AS service_point_count\nFROM\n `data-ge.water_meter_info`\nGROUP BY\n `${dimension_column}`\nORDER BY\n service_point_count DESC\nLIMIT ${top_n};"
},
"applicability": {
"constraints": {
"notes": [
"维度 `station` 基数较高 (36)，建议 Top-N 查询时结合业务场景合理设置 N 值。"
],
"fk_join_available": false,
"dim_cardinality_hint": 36
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "用水点数：对 `service_point_id` 进行去重计数。排名依据为各维度分类下的用水点总数。统计粒度为“指定维度”。"
},
{
"id": "snpt_ratio-service-points-by-dimension",
"desc": "计算在指定维度下，各分类的用水点数占总用水点数的百分比，以分析其分布构成。",
"type": "ratio",
"title": "各维度用水点数占比",
"examples": [
"不同水表类型（meter_type）的分布情况",
"各个区域的用水点占比是多少"
],
"variables": [
{
"name": "dimension_column",
"type": "column",
"default": "meter_type"
}
],
"dialect_sql": {
"mysql": "SELECT\n `${dimension_column}`,\n COUNT(DISTINCT service_point_id) AS service_point_count,\n COUNT(DISTINCT service_point_id) * 100.0 / SUM(COUNT(DISTINCT service_point_id)) OVER () AS percentage\nFROM\n `data-ge.water_meter_info`\nGROUP BY\n `${dimension_column}`\nORDER BY\n service_point_count DESC;"
},
"applicability": {
"constraints": {
"notes": [
"SQL模板使用了窗口函数 SUM() OVER()，请确保MySQL版本支持（8.0+）。"
],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "用水点数占比：某分类下的用水点数 / 总用水点数。用水点数以 `service_point_id` 去重计数。统计粒度为“指定维度”。"
},
{
"id": "snpt_quality-check-duplicate-spid",
"desc": "查找在用水点信息表中存在重复的 `service_point_id`，用于数据质量校验。",
"type": "quality",
"title": "检查重复的用水点ID",
"examples": [
"检查是否存在重复的水表档案",
"校验用水点ID的唯一性"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT\n service_point_id,\n COUNT(*) AS occurrences\nFROM\n `data-ge.water_meter_info`\nGROUP BY\n service_point_id\nHAVING\n COUNT(*) > 1;"
},
"applicability": {
"constraints": {
"notes": [
"预期返回结果为空。若有返回，则表示数据存在一致性问题，`service_point_id` 未能作为唯一主键。"
],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "重复项：指 `service_point_id` 出现次数大于1的记录。此ID应为表的主键，理论上不应重复。"
},
{
"id": "snpt_sample-filter-service-points-by-dims",
"desc": "根据区域、水表类型、供水所等多个维度组合条件，筛选出符合条件的用水点明细。",
"type": "sample",
"title": "多维度筛选用水点列表",
"examples": [
"查询城区的机械表有哪些",
"拉取某个供水所下特定口径水表的列表"
],
"variables": [
{
"name": "district_name",
"type": "string",
"default": "城区"
},
{
"name": "meter_type_name",
"type": "string",
"default": "机械表"
},
{
"name": "limit_num",
"type": "int",
"default": 100
}
],
"dialect_sql": {
"mysql": "SELECT\n service_point_id,\n account_id,\n district,\n supply_office,\n meter_type,\n meter_subtype,\n meter_diameter\nFROM\n `data-ge.water_meter_info`\nWHERE\n district = '${district_name}'\n AND meter_type = '${meter_type_name}'\n -- AND meter_status = '有效' -- 可选：根据画像，该列为常量'有效'，可不加\nLIMIT ${limit_num};"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": null,
"required_columns": [
"service_point_id",
"account_id",
"district",
"supply_office",
"meter_type",
"meter_subtype",
"meter_diameter"
]
},
"business_caliber": "返回满足所有筛选条件的用水点明细信息。`meter_status` 列只有一个值 '有效'，通常无需作为筛选条件。"
}
]

230
demo/水务/水务-gpt5-ge-desc.json Normal file
View File

@ -0,0 +1,230 @@
{
"role": "dimension",
"time": {
"range": null,
"column": null,
"has_gaps": null,
"granularity": "unknown"
},
"grain": [
"service_point_id"
],
"table": "data-ge.water_meter_info",
"columns": [
{
"name": "supply_office",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；11 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 11,
"distinct_ratio": 0.03666666666666667,
"pk_candidate_score": 0.05,
"metric_candidate_score": 0.0
},
{
"name": "station",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；36 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 36,
"distinct_ratio": 0.12,
"pk_candidate_score": 0.1,
"metric_candidate_score": 0.0
},
{
"name": "district",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；13 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 13,
"distinct_ratio": 0.043333333333333335,
"pk_candidate_score": 0.05,
"metric_candidate_score": 0.0
},
{
"name": "meter_diameter",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；8 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 8,
"distinct_ratio": 0.02666666666666667,
"pk_candidate_score": 0.03,
"metric_candidate_score": 0.0
},
{
"name": "meter_status",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；单一取值（\"有效\"）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 1,
"distinct_ratio": 0.0033333333333333335,
"pk_candidate_score": 0.0,
"metric_candidate_score": 0.0
},
{
"name": "meter_subtype",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；9 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 9,
"distinct_ratio": 0.03,
"pk_candidate_score": 0.03,
"metric_candidate_score": 0.0
},
{
"name": "meter_type",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；5 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 5,
"distinct_ratio": 0.016666666666666666,
"pk_candidate_score": 0.02,
"metric_candidate_score": 0.0
},
{
"name": "installation_position",
"dtype": "string",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "非空；4 个枚举值（GE 约束）",
"enumish": true,
"null_rate": 0.0,
"top_values": [],
"semantic_type": "dimension",
"distinct_count": 4,
"distinct_ratio": 0.013333333333333334,
"pk_candidate_score": 0.02,
"metric_candidate_score": 0.0
},
{
"name": "service_point_id",
"dtype": "unknown",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "命名指示标识列；未提供唯一性或非空验证",
"enumish": null,
"null_rate": null,
"top_values": [],
"semantic_type": "id",
"distinct_count": null,
"distinct_ratio": null,
"pk_candidate_score": 0.6,
"metric_candidate_score": 0.05
},
{
"name": "account_id",
"dtype": "unknown",
"stats": {
"max": null,
"min": null,
"std": null,
"mean": null,
"skewness": null
},
"comment": "命名指示账户标识；未提供唯一性或非空验证",
"enumish": null,
"null_rate": null,
"top_values": [],
"semantic_type": "id",
"distinct_count": null,
"distinct_ratio": null,
"pk_candidate_score": 0.5,
"metric_candidate_score": 0.05
}
],
"quality": {
"warning_hints": [
"以下列未设置非空校验：service_point_id, account_id（空值情况未知）",
"未识别到时间列"
],
"failed_expectations": []
},
"row_count": 300,
"fk_candidates": [],
"confidence_notes": [
"role 判定为 dimension：表内列均为枚举/分类或ID，未发现数值型度量或时间列；34/34 期望均为分类枚举/非空与去重比例。",
"grain 猜测为 service_point_id：仅依据命名启发式，缺少唯一性与非空度量佐证（置信度较低）。",
"未识别时间列：列名与期望均未涉及日期/时间，也无最小/最大时间范围可推断。"
],
"primary_key_candidates": []
}

372
demo/水务/水务-gpt5-snippet-alias.json Normal file
View File

@ -0,0 +1,372 @@
[
{
"id": "snpt_topn_station",
"aliases": [
{
"text": "站点水表排行前N",
"tone": "中性"
},
{
"text": "哪个站点表最多",
"tone": "口语"
},
{
"text": "按站点水表TopN",
"tone": "专业"
}
],
"keywords": [
"TopN",
"排名",
"排行",
"station",
"站点",
"水表数",
"meter count",
"distinct",
"去重",
"聚合",
"排序",
"榜单"
],
"intent_tags": [
"topn",
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_share_district",
"aliases": [
{
"text": "各辖区水表占比",
"tone": "中性"
},
{
"text": "哪个辖区占比高",
"tone": "口语"
},
{
"text": "按辖区水表比例",
"tone": "专业"
}
],
"keywords": [
"占比",
"ratio",
"district",
"辖区",
"水表数",
"meter count",
"distinct",
"去重",
"百分比",
"份额",
"聚合",
"排序",
"分布"
],
"intent_tags": [
"ratio",
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_dist_diameter",
"aliases": [
{
"text": "表径水表数分布",
"tone": "中性"
},
{
"text": "不同口径有多少",
"tone": "口语"
},
{
"text": "按表径去重计数",
"tone": "专业"
}
],
"keywords": [
"分布",
"distribution",
"meter_diameter",
"表径",
"水表数",
"meter count",
"distinct",
"去重",
"聚合",
"类别",
"category",
"条形图",
"饼图",
"排行"
],
"intent_tags": [
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_type_subtype_matrix",
"aliases": [
{
"text": "类型×子类水表数",
"tone": "中性"
},
{
"text": "看各类型各子类",
"tone": "口语"
},
{
"text": "类型子类组合统计",
"tone": "专业"
}
],
"keywords": [
"类型",
"type",
"子类",
"subtype",
"组合",
"matrix",
"交叉分析",
"cross-tab",
"水表数",
"meter count",
"distinct",
"去重",
"分布",
"聚合",
"维度"
],
"intent_tags": [
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_quality_spid_uniq",
"aliases": [
{
"text": "服务点ID唯一性检",
"tone": "专业"
},
{
"text": "服务点ID有重复吗",
"tone": "口语"
},
{
"text": "服务点ID完整性评估",
"tone": "中性"
}
],
"keywords": [
"质量检查",
"quality",
"唯一性",
"uniqueness",
"重复",
"duplicate",
"空值",
"NULL",
"完整性",
"integrity",
"service_point_id",
"数据质量",
"统计",
"去重",
"异常检测"
],
"intent_tags": [
"quality"
]
},
{
"id": "snpt_quality_account_nulls",
"aliases": [
{
"text": "账户ID缺失明细",
"tone": "中性"
},
{
"text": "看看哪些账户为空",
"tone": "口语"
},
{
"text": "account_id空值样本",
"tone": "专业"
}
],
"keywords": [
"质量检查",
"缺失",
"missing",
"空值",
"NULL",
"account_id",
"样本",
"sample",
"抽样",
"sampling",
"明细",
"排查",
"过滤",
"WHERE",
"LIMIT"
],
"intent_tags": [
"quality",
"sample"
]
},
{
"id": "snpt_sample_random_rows",
"aliases": [
{
"text": "随机抽样水表明细",
"tone": "中性"
},
{
"text": "随机取几条看看",
"tone": "口语"
},
{
"text": "RAND()样本抽取",
"tone": "专业"
}
],
"keywords": [
"随机",
"random",
"样本",
"sample",
"抽样",
"sampling",
"明细",
"details",
"质检",
"QA",
"RAND()",
"LIMIT",
"抽取",
"数据验证"
],
"intent_tags": [
"sample"
]
},
{
"id": "snpt_filter_office_type_where",
"aliases": [
{
"text": "按所与类型过滤有效",
"tone": "专业"
},
{
"text": "筛选某所的指定类型",
"tone": "中性"
},
{
"text": "只看这所的这种表",
"tone": "口语"
}
],
"keywords": [
"过滤",
"filter",
"WHERE",
"supply_office",
"营业所",
"meter_type",
"类型",
"meter_status",
"有效",
"条件片段",
"筛选",
"查询拼接",
"字段",
"约束"
],
"intent_tags": [
"filter"
]
},
{
"id": "snpt_office_station_dist",
"aliases": [
{
"text": "所站组合水表数",
"tone": "中性"
},
{
"text": "各站在各所有多少",
"tone": "口语"
},
{
"text": "营业所×站点分布",
"tone": "专业"
}
],
"keywords": [
"supply_office",
"营业所",
"station",
"站点",
"层级",
"hierarchy",
"分布",
"distribution",
"水表数",
"meter count",
"distinct",
"去重",
"聚合",
"交叉分析",
"排行"
],
"intent_tags": [
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_total_meter_baseline",
"aliases": [
{
"text": "水表总量基线",
"tone": "中性"
},
{
"text": "现在有多少水表",
"tone": "口语"
},
{
"text": "全表去重总数",
"tone": "专业"
}
],
"keywords": [
"总量",
"total",
"baseline",
"基线",
"水表总数",
"meter total",
"service_point_id",
"distinct",
"去重",
"分母",
"denominator",
"占比",
"聚合",
"汇总",
"snapshot"
],
"intent_tags": [
"aggregate"
]
}
]

330
demo/水务/水务-gpt5-snippet.json Normal file
View File

@ -0,0 +1,330 @@
[
{
"id": "snpt_topn_station",
"desc": "按站点统计水表数量并取前N",
"type": "topn",
"title": "站点TopN水表数",
"examples": [
"各站点水表数量排名前10",
"站点水表覆盖情况排行"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 10
}
],
"dialect_sql": {
"mysql": "SELECT station,\n COUNT(DISTINCT service_point_id) AS meter_cnt\nFROM `data-ge`.`water_meter_info`\nGROUP BY station\nORDER BY meter_cnt DESC\nLIMIT {{top_n}};"
},
"applicability": {
"constraints": {
"notes": [
"TopN建议N<=36",
"以service_point_id去重计数",
"无时间列，无法做趋势"
],
"fk_join_available": false,
"dim_cardinality_hint": 36
},
"time_column": null,
"required_columns": [
"station",
"service_point_id"
]
},
"business_caliber": "水表数=按service_point_id去重计数；粒度=站点。仅统计当前表中的有效记录（不含时间口径）。安全限制：用于分析排名，避免扩大LIMIT造成全量导出。"
},
{
"id": "snpt_share_district",
"desc": "统计各辖区水表数及其占比",
"type": "ratio",
"title": "辖区水表占比",
"examples": [
"各辖区水表占比",
"哪个辖区水表最多"
],
"variables": [],
"dialect_sql": {
"mysql": "WITH by_district AS (\n SELECT district, COUNT(DISTINCT service_point_id) AS meter_cnt\n FROM `data-ge`.`water_meter_info`\n GROUP BY district\n), tot AS (\n SELECT COUNT(DISTINCT service_point_id) AS total_cnt\n FROM `data-ge`.`water_meter_info`\n)\nSELECT b.district,\n b.meter_cnt,\n ROUND(b.meter_cnt / NULLIF(t.total_cnt, 0) * 100, 2) AS pct\nFROM by_district b\nCROSS JOIN tot t\nORDER BY pct DESC, b.district;"
},
"applicability": {
"constraints": {
"notes": [
"占比分母为全表service_point_id去重总数",
"service_point_id为空将被忽略"
],
"fk_join_available": false,
"dim_cardinality_hint": 13
},
"time_column": null,
"required_columns": [
"district",
"service_point_id"
]
},
"business_caliber": "水表数=按service_point_id去重计数；粒度=辖区。占比=辖区水表数/全表水表总数。安全限制：仅基于本表，不代表全市/全网口径；无时间维度。"
},
{
"id": "snpt_dist_diameter",
"desc": "按表径统计水表数量分布",
"type": "aggregate",
"title": "表径分布统计",
"examples": [
"不同口径水表有多少",
"查看表径分布情况"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT meter_diameter,\n COUNT(DISTINCT service_point_id) AS meter_cnt\nFROM `data-ge`.`water_meter_info`\nGROUP BY meter_diameter\nORDER BY meter_cnt DESC, meter_diameter;"
},
"applicability": {
"constraints": {
"notes": [
"以service_point_id去重计数",
"适合绘制条形图/饼图"
],
"fk_join_available": false,
"dim_cardinality_hint": 8
},
"time_column": null,
"required_columns": [
"meter_diameter",
"service_point_id"
]
},
"business_caliber": "水表数=按service_point_id去重计数；粒度=表径。安全限制：仅用于分布分析，不含时间过滤；避免用于明细导出。"
},
{
"id": "snpt_type_subtype_matrix",
"desc": "统计水表类型与子类组合的数量",
"type": "aggregate",
"title": "类型子类分布",
"examples": [
"不同类型与子类的水表数量",
"查看类型与子类的组合分布"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT meter_type,\n meter_subtype,\n COUNT(DISTINCT service_point_id) AS meter_cnt\nFROM `data-ge`.`water_meter_info`\nGROUP BY meter_type, meter_subtype\nORDER BY meter_cnt DESC, meter_type, meter_subtype;"
},
"applicability": {
"constraints": {
"notes": [
"组合基数<=5×9=45",
"以service_point_id去重计数"
],
"fk_join_available": false,
"dim_cardinality_hint": 45
},
"time_column": null,
"required_columns": [
"meter_type",
"meter_subtype",
"service_point_id"
]
},
"business_caliber": "水表数=按service_point_id去重计数；粒度=类型×子类组合。安全限制：仅用于汇总分析，不包含时间或业务状态变化。"
},
{
"id": "snpt_quality_spid_uniq",
"desc": "评估service_point_id的空值与重复情况",
"type": "quality",
"title": "服务点唯一性检",
"examples": [
"检查服务点ID是否唯一",
"统计service_point_id空值与重复情况"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT\n COUNT(*) AS total_rows,\n SUM(service_point_id IS NULL) AS null_cnt,\n COUNT(DISTINCT service_point_id) AS distinct_cnt,\n (COUNT(*) - COUNT(DISTINCT service_point_id)) AS duplicate_rows_est,\n (\n SELECT COUNT(*) FROM (\n SELECT service_point_id\n FROM `data-ge`.`water_meter_info`\n GROUP BY service_point_id\n HAVING COUNT(*) > 1\n ) AS dup\n ) AS dup_key_groups\nFROM `data-ge`.`water_meter_info`;"
},
"applicability": {
"constraints": {
"notes": [
"用于键完整性检查",
"重复行估算=总行数-去重数"
],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "质量检查口径：在本表内评估service_point_id的非空与唯一性，不代表跨表全局唯一。安全限制：仅输出汇总指标，不暴露明细重复值。"
},
{
"id": "snpt_quality_account_nulls",
"desc": "抽取account_id为空的记录用于排查",
"type": "quality",
"title": "账户ID缺失明细",
"examples": [
"列出account_id为空的水表",
"抽样查看账户缺失的数据行"
],
"variables": [
{
"name": "limit_n",
"type": "int",
"default": 50
}
],
"dialect_sql": {
"mysql": "SELECT *\nFROM `data-ge`.`water_meter_info`\nWHERE account_id IS NULL\nLIMIT {{limit_n}};"
},
"applicability": {
"constraints": {
"notes": [
"明细仅限小样本抽取",
"建议LIMIT<=100，避免全量导出"
],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": null,
"required_columns": [
"account_id"
]
},
"business_caliber": "质量抽样：筛出账户ID缺失的水表记录，便于核对。安全限制：仅用于样本排查，不建议在生产中全量导出；如需口径统计请改为COUNT聚合。"
},
{
"id": "snpt_sample_random_rows",
"desc": "随机抽取水表信息用于人工核验",
"type": "sample",
"title": "随机抽样明细",
"examples": [
"抽样查看水表信息",
"随机抽取20条做质检"
],
"variables": [
{
"name": "sample_size",
"type": "int",
"default": 20
}
],
"dialect_sql": {
"mysql": "SELECT *\nFROM `data-ge`.`water_meter_info`\nORDER BY RAND()\nLIMIT {{sample_size}};"
},
"applicability": {
"constraints": {
"notes": [
"使用RAND()随机，样本不可复现",
"建议限制样本量"
],
"fk_join_available": false,
"dim_cardinality_hint": 300
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "样本抽取：从本表随机返回若干行明细。安全限制：避免扩大LIMIT进行全量下载；如需可复现样本，请改用带种子的随机方法（MySQL不原生支持）。"
},
{
"id": "snpt_filter_office_type_where",
"desc": "常用WHERE筛选条件片段：按营业所与类型且为有效",
"type": "sample",
"title": "机构类型筛选片",
"examples": [
"筛选A营业所的机械表",
"仅查看某营业所的指定类型水表"
],
"variables": [
{
"name": "supply_office",
"type": "string"
},
{
"name": "meter_type",
"type": "string"
}
],
"dialect_sql": {
"mysql": "WHERE supply_office = '{{supply_office}}'\n AND meter_type = '{{meter_type}}'\n AND meter_status = '有效'"
},
"applicability": {
"constraints": {
"notes": [
"这是条件片段，可拼接到其他查询",
"meter_status当前为单一值“有效”"
],
"fk_join_available": false,
"dim_cardinality_hint": 11
},
"time_column": null,
"required_columns": [
"supply_office",
"meter_type",
"meter_status"
]
},
"business_caliber": "过滤口径：仅保留指定营业所与指定水表类型、且状态为“有效”的记录。安全限制：为片段用途，需拼接在SELECT…FROM之后使用。"
},
{
"id": "snpt_office_station_dist",
"desc": "按营业所与站点组合统计水表数",
"type": "aggregate",
"title": "所站层级分布",
"examples": [
"按营业所查看各站点水表数",
"所站两级的水表分布情况"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT supply_office,\n station,\n COUNT(DISTINCT service_point_id) AS meter_cnt\nFROM `data-ge`.`water_meter_info`\nGROUP BY supply_office, station\nORDER BY supply_office, meter_cnt DESC, station;"
},
"applicability": {
"constraints": {
"notes": [
"组合基数<=11×36=396",
"以service_point_id去重计数",
"如结果过长可再按TopN筛选"
],
"fk_join_available": false,
"dim_cardinality_hint": 396
},
"time_column": null,
"required_columns": [
"supply_office",
"station",
"service_point_id"
]
},
"business_caliber": "水表数=按service_point_id去重计数；粒度=营业所×站点。安全限制：结果行数可能较多，建议在可视化端增加筛选或分页。"
},
{
"id": "snpt_total_meter_baseline",
"desc": "获取全表水表去重总量基线",
"type": "aggregate",
"title": "水表总量基线",
"examples": [
"当前有多少只水表",
"作为占比分析的分母基线"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT COUNT(DISTINCT service_point_id) AS meter_total\nFROM `data-ge`.`water_meter_info`;"
},
"applicability": {
"constraints": {
"notes": [
"作为其他占比/分摊分母基线",
"忽略service_point_id为空的记录"
],
"fk_join_available": false,
"dim_cardinality_hint": 300
},
"time_column": null,
"required_columns": [
"service_point_id"
]
},
"business_caliber": "水表总量=按service_point_id去重计数；基于当前表的全量记录。安全限制：无时间维度，无法反映存量随时间变化。"
}
]

415
demo/水务/水务-qwen3-coder-480b-ge-desc.json Normal file
View File

@ -0,0 +1,415 @@
{
"role": "dimension",
"time": {
"range": null,
"column": null,
"has_gaps": null,
"granularity": "unknown"
},
"grain": [
"account_id",
"service_point_id"
],
"table": "data-ge.water_meter_info",
"columns": [
{
"name": "supply_office",
"dtype": "string",
"stats": {},
"comment": "供水管理所名称，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "宝山供水管理所"
},
{
"pct": null,
"value": "黄浦供水管理所"
},
{
"pct": null,
"value": "青东供水管理所"
},
{
"pct": null,
"value": "虹口供水管理所"
},
{
"pct": null,
"value": "闸北供水管理所"
},
{
"pct": null,
"value": "松北供水管理所"
},
{
"pct": null,
"value": "杨浦供水管理所"
},
{
"pct": null,
"value": "长宁供水管理所"
},
{
"pct": null,
"value": "闵行供水管理所"
},
{
"pct": null,
"value": "徐汇供水管理所"
},
{
"pct": null,
"value": "普陀供水管理所"
}
],
"semantic_type": "dimension",
"distinct_count": 11,
"distinct_ratio": 0.03666666666666667,
"pk_candidate_score": 0.11,
"metric_candidate_score": 0.0
},
{
"name": "station",
"dtype": "string",
"stats": {},
"comment": "站点名称，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "新闸站"
},
{
"pct": null,
"value": "宝杨站"
},
{
"pct": null,
"value": "江川站"
},
{
"pct": null,
"value": "长江站"
},
{
"pct": null,
"value": "市光站"
},
{
"pct": null,
"value": "徐泾站"
},
{
"pct": null,
"value": "真北站"
},
{
"pct": null,
"value": "半淞园站"
},
{
"pct": null,
"value": "芙蓉江站"
},
{
"pct": null,
"value": "密云站"
}
],
"semantic_type": "dimension",
"distinct_count": 36,
"distinct_ratio": 0.12,
"pk_candidate_score": 0.36,
"metric_candidate_score": 0.0
},
{
"name": "district",
"dtype": "string",
"stats": {},
"comment": "行政区划名称，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "普陀区"
},
{
"pct": null,
"value": "闵行区"
},
{
"pct": null,
"value": "嘉定区"
},
{
"pct": null,
"value": "杨浦区"
},
{
"pct": null,
"value": "徐汇区"
},
{
"pct": null,
"value": "黄浦区"
},
{
"pct": null,
"value": "松江区"
},
{
"pct": null,
"value": "长宁区"
},
{
"pct": null,
"value": "青浦区"
},
{
"pct": null,
"value": "虹口区"
}
],
"semantic_type": "dimension",
"distinct_count": 13,
"distinct_ratio": 0.043333333333333335,
"pk_candidate_score": 0.13,
"metric_candidate_score": 0.0
},
{
"name": "meter_diameter",
"dtype": "string",
"stats": {},
"comment": "水表直径规格，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "20mm"
},
{
"pct": null,
"value": "15mm"
},
{
"pct": null,
"value": "25mm"
},
{
"pct": null,
"value": "40mm"
},
{
"pct": null,
"value": "150mm"
},
{
"pct": null,
"value": "100mm"
},
{
"pct": null,
"value": "80mm"
},
{
"pct": null,
"value": "50mm"
}
],
"semantic_type": "dimension",
"distinct_count": 8,
"distinct_ratio": 0.02666666666666667,
"pk_candidate_score": 0.08,
"metric_candidate_score": 0.0
},
{
"name": "meter_status",
"dtype": "string",
"stats": {},
"comment": "水表状态，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "有效"
}
],
"semantic_type": "dimension",
"distinct_count": 1,
"distinct_ratio": 0.0033333333333333335,
"pk_candidate_score": 0.01,
"metric_candidate_score": 0.0
},
{
"name": "meter_subtype",
"dtype": "string",
"stats": {},
"comment": "水表子类型，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "旋翼半液封式"
},
{
"pct": null,
"value": "超声波式"
},
{
"pct": null,
"value": "旋翼湿式（指针式）"
},
{
"pct": null,
"value": "旋翼湿式（数字指针式）"
},
{
"pct": null,
"value": "电磁式"
},
{
"pct": null,
"value": "无直管段要求超声波式"
},
{
"pct": null,
"value": "无直管段要求电磁式"
},
{
"pct": null,
"value": "垂直螺翼干式"
},
{
"pct": null,
"value": "机械容积式"
}
],
"semantic_type": "dimension",
"distinct_count": 9,
"distinct_ratio": 0.03,
"pk_candidate_score": 0.09,
"metric_candidate_score": 0.0
},
{
"name": "meter_type",
"dtype": "string",
"stats": {},
"comment": "水表类型，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "容积式机械水表"
},
{
"pct": null,
"value": "速度式机械水表"
},
{
"pct": null,
"value": "电磁式远传水表"
},
{
"pct": null,
"value": "速度式机电远传水表"
},
{
"pct": null,
"value": "超声波式远传水表"
}
],
"semantic_type": "dimension",
"distinct_count": 5,
"distinct_ratio": 0.016666666666666666,
"pk_candidate_score": 0.05,
"metric_candidate_score": 0.0
},
{
"name": "installation_position",
"dtype": "string",
"stats": {},
"comment": "安装位置，枚举值",
"enumish": true,
"null_rate": 0.0,
"top_values": [
{
"pct": null,
"value": "嵌墙表"
},
{
"pct": null,
"value": "管道井表"
},
{
"pct": null,
"value": "地下表"
},
{
"pct": null,
"value": "龙头表"
}
],
"semantic_type": "dimension",
"distinct_count": 4,
"distinct_ratio": 0.013333333333333334,
"pk_candidate_score": 0.04,
"metric_candidate_score": 0.0
},
{
"name": "account_id",
"dtype": "string",
"stats": {},
"comment": "账户ID",
"enumish": false,
"null_rate": null,
"top_values": [],
"semantic_type": "id",
"distinct_count": null,
"distinct_ratio": null,
"pk_candidate_score": 0.95,
"metric_candidate_score": 0.0
},
{
"name": "service_point_id",
"dtype": "string",
"stats": {},
"comment": "服务点ID",
"enumish": false,
"null_rate": null,
"top_values": [],
"semantic_type": "id",
"distinct_count": null,
"distinct_ratio": null,
"pk_candidate_score": 0.95,
"metric_candidate_score": 0.0
}
],
"quality": {
"warning_hints": [],
"failed_expectations": []
},
"row_count": 300,
"fk_candidates": [],
"confidence_notes": [
"role判定为dimension，因所有列均为枚举或ID类型，无metric列",
"grain依据account_id和service_point_id为唯一标识推测",
"未发现时间列，因此time字段为null"
],
"primary_key_candidates": [
[
"account_id"
],
[
"service_point_id"
]
]
}

286
demo/水务/水务-qwen3-coder-480b-snippet-alias.json Normal file
View File

@ -0,0 +1,286 @@
[
{
"id": "snpt_water_meter_top_supply_office",
"aliases": [
{
"text": "供水所水表排行",
"tone": "中性"
},
{
"text": "哪个供水所水表最多",
"tone": "口语"
},
{
"text": "供水管理所水表TopN统计",
"tone": "专业"
}
],
"keywords": [
"水表",
"供水管理所",
"排行",
"TopN",
"数量",
"统计",
"count",
"排名",
"前N",
"供水所",
"水表数",
"维度聚合",
"by_dimension",
"topn"
],
"intent_tags": [
"topn",
"by_dimension"
]
},
{
"id": "snpt_water_meter_top_station",
"aliases": [
{
"text": "站点水表数量排行",
"tone": "中性"
},
{
"text": "哪个站点水表最多",
"tone": "口语"
},
{
"text": "站点维度水表TopN分析",
"tone": "专业"
}
],
"keywords": [
"水表",
"站点",
"排行",
"TopN",
"数量",
"统计",
"count",
"排名",
"前N",
"站点数",
"维度聚合",
"by_dimension",
"topn"
],
"intent_tags": [
"topn",
"by_dimension"
]
},
{
"id": "snpt_water_meter_top_district",
"aliases": [
{
"text": "区域水表数量排名",
"tone": "中性"
},
{
"text": "哪个区水表最多",
"tone": "口语"
},
{
"text": "行政区水表TopN统计",
"tone": "专业"
}
],
"keywords": [
"水表",
"区域",
"行政区",
"排行",
"TopN",
"数量",
"统计",
"count",
"排名",
"前N",
"区",
"水表数",
"维度聚合",
"by_dimension",
"topn"
],
"intent_tags": [
"topn",
"by_dimension"
]
},
{
"id": "snpt_water_meter_share_by_type",
"aliases": [
{
"text": "水表类型占比",
"tone": "中性"
},
{
"text": "哪种水表用得最多",
"tone": "口语"
},
{
"text": "水表类型分布比例",
"tone": "专业"
}
],
"keywords": [
"水表",
"类型",
"占比",
"比例",
"ratio",
"分布",
"meter_type",
"百分比",
"分类统计",
"水表类型",
"ratio",
"aggregate",
"by_dimension"
],
"intent_tags": [
"ratio",
"by_dimension"
]
},
{
"id": "snpt_water_meter_subtype_distribution",
"aliases": [
{
"text": "水表子类型分布",
"tone": "中性"
},
{
"text": "各种子类型水表情况",
"tone": "口语"
},
{
"text": "水表子类型计数与占比",
"tone": "专业"
}
],
"keywords": [
"水表",
"子类型",
"分布",
"数量",
"占比",
"meter_subtype",
"统计",
"count",
"百分比",
"分类统计",
"aggregate",
"by_dimension"
],
"intent_tags": [
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_water_meter_installation_position_stats",
"aliases": [
{
"text": "安装位置统计",
"tone": "中性"
},
{
"text": "哪种位置装表最多",
"tone": "口语"
},
{
"text": "水表安装位置分布",
"tone": "专业"
}
],
"keywords": [
"水表",
"安装位置",
"统计",
"分布",
"installation_position",
"数量",
"count",
"位置",
"安装点",
"aggregate",
"by_dimension"
],
"intent_tags": [
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_water_meter_grain_check",
"aliases": [
{
"text": "主键粒度校验",
"tone": "中性"
},
{
"text": "数据有没有重复",
"tone": "口语"
},
{
"text": "数据粒度一致性检查",
"tone": "专业"
}
],
"keywords": [
"主键",
"粒度",
"校验",
"质量",
"quality",
"重复",
"唯一性",
"account_id",
"service_point_id",
"数据校验",
"质量检查",
"异常检测"
],
"intent_tags": [
"quality"
]
},
{
"id": "snpt_water_meter_sample_records",
"aliases": [
{
"text": "水表数据抽样",
"tone": "中性"
},
{
"text": "给我看点水表数据",
"tone": "口语"
},
{
"text": "水表记录样本抽取",
"tone": "专业"
}
],
"keywords": [
"水表",
"样本",
"抽样",
"sample",
"随机",
"记录",
"抽查",
"limit",
"数据结构",
"数据示例",
"sample",
"limit_rows"
],
"intent_tags": [
"sample"
]
}
]

235
demo/水务/水务-qwen3-coder-480b-snippet.json Normal file
View File

@ -0,0 +1,235 @@
[
{
"id": "snpt_water_meter_top_supply_office",
"desc": "统计各供水管理所下辖水表数量并排序",
"type": "topn",
"title": "供水管理所水表数量排行",
"examples": [
"列出水表最多的前10个供水管理所",
"各供水所水表数量排名"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 10
}
],
"dialect_sql": {
"mysql": "SELECT supply_office AS dim_value, COUNT(*) AS metric_value FROM `data-ge.water_meter_info` GROUP BY supply_office ORDER BY metric_value DESC LIMIT {{top_n}}"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": 11
},
"time_column": "nullable",
"required_columns": [
"supply_office"
]
},
"business_caliber": "按供水管理所维度聚合水表总数，粒度=供水管理所"
},
{
"id": "snpt_water_meter_top_station",
"desc": "统计各个站点下辖水表数量并排序",
"type": "topn",
"title": "站点水表数量排行",
"examples": [
"列出水表最多的前10个站点",
"各站点水表数量排名"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 10
}
],
"dialect_sql": {
"mysql": "SELECT station AS dim_value, COUNT(*) AS metric_value FROM `data-ge.water_meter_info` GROUP BY station ORDER BY metric_value DESC LIMIT {{top_n}}"
},
"applicability": {
"constraints": {
"notes": [
"高基数维度建议LIMIT<=50"
],
"fk_join_available": false,
"dim_cardinality_hint": 36
},
"time_column": "nullable",
"required_columns": [
"station"
]
},
"business_caliber": "按站点维度聚合水表总数，粒度=站点"
},
{
"id": "snpt_water_meter_top_district",
"desc": "统计各区水表数量并排序",
"type": "topn",
"title": "区域水表数量排行",
"examples": [
"列出各区水表数量排名",
"哪个区的水表最多？"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 10
}
],
"dialect_sql": {
"mysql": "SELECT district AS dim_value, COUNT(*) AS metric_value FROM `data-ge.water_meter_info` GROUP BY district ORDER BY metric_value DESC LIMIT {{top_n}}"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": 13
},
"time_column": "nullable",
"required_columns": [
"district"
]
},
"business_caliber": "按行政区划维度聚合水表总数，粒度=区"
},
{
"id": "snpt_water_meter_share_by_type",
"desc": "计算各类水表占总水表的比例",
"type": "ratio",
"title": "水表类型占比分布",
"examples": [
"各类水表占比是多少？",
"哪种类型的水表使用最广泛？"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT meter_type AS dim_value, COUNT(*) * 100.0 / (SELECT COUNT(*) FROM `data-ge.water_meter_info`) AS ratio_percent FROM `data-ge.water_meter_info` GROUP BY meter_type ORDER BY ratio_percent DESC"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": 5
},
"time_column": "nullable",
"required_columns": [
"meter_type"
]
},
"business_caliber": "按水表类型分类计算其占比，粒度=水表类型"
},
{
"id": "snpt_water_meter_subtype_distribution",
"desc": "展示不同水表子类型的数量及比例",
"type": "aggregate",
"title": "水表子类型分布情况",
"examples": [
"各种子类型水表的数量和占比",
"哪种子类型水表最多？"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT meter_subtype AS dim_value, COUNT(*) AS count_value, ROUND(COUNT(*) * 100.0 / (SELECT COUNT(*) FROM `data-ge.water_meter_info`), 2) AS percentage FROM `data-ge.water_meter_info` GROUP BY meter_subtype ORDER BY count_value DESC"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": 9
},
"time_column": "nullable",
"required_columns": [
"meter_subtype"
]
},
"business_caliber": "按水表子类型进行计数和百分比统计，粒度=水表子类型"
},
{
"id": "snpt_water_meter_installation_position_stats",
"desc": "统计不同安装位置下的水表数量",
"type": "aggregate",
"title": "安装位置分布统计",
"examples": [
"各种安装位置的水表数量",
"哪种安装位置最为常见？"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT installation_position AS dim_value, COUNT(*) AS count_value FROM `data-ge.water_meter_info` GROUP BY installation_position ORDER BY count_value DESC"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": 4
},
"time_column": "nullable",
"required_columns": [
"installation_position"
]
},
"business_caliber": "按安装位置对水表进行分组计数，粒度=安装位置"
},
{
"id": "snpt_water_meter_grain_check",
"desc": "验证 account_id 和 service_point_id 是否构成唯一组合",
"type": "quality",
"title": "主键粒度校验",
"examples": [
"这张表的数据粒度是否正确？",
"是否存在重复的服务点记录？"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT IF(COUNT(*) = COUNT(DISTINCT account_id, service_point_id), 'PASS', 'FAIL') AS grain_check_result FROM `data-ge.water_meter_info`"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": "nullable",
"required_columns": [
"account_id",
"service_point_id"
]
},
"business_caliber": "检验数据是否符合预期的主键粒度（account_id + service_point_id）"
},
{
"id": "snpt_water_meter_sample_records",
"desc": "随机抽取部分水表信息用于查看结构",
"type": "sample",
"title": "样本抽取",
"examples": [
"给我看几条水表数据的例子",
"抽查一些原始数据看看格式"
],
"variables": [
{
"name": "limit_rows",
"type": "int",
"default": 5
}
],
"dialect_sql": {
"mysql": "SELECT * FROM `data-ge.water_meter_info` ORDER BY RAND() LIMIT {{limit_rows}}"
},
"applicability": {
"constraints": {
"notes": [],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": "nullable",
"required_columns": []
},
"business_caliber": "从全量数据中随机采样若干条记录供参考"
}
]

1
demo/水务/水务-qwen3-next-80b-ge-desc.json Normal file
View File

File diff suppressed because one or more lines are too long

249
demo/水务/水务-qwen3-next-80b-snippet-alias.json Normal file
View File

@ -0,0 +1,249 @@
[
{
"id": "snpt_topn_supply_office_by_account",
"aliases": [
{
"text": "哪个供水所用户最多？",
"tone": "口语"
},
{
"text": "按供应办公室统计账户数量",
"tone": "中性"
},
{
"text": "供应办公室账户数TopN排名",
"tone": "专业"
}
],
"keywords": [
"供应办公室",
"账户数",
"TopN",
"排行",
"统计",
"account_id",
"supply_office",
"去重",
"高占比",
"维度分析",
"by_dimension",
"aggregate",
"topn"
],
"intent_tags": [
"topn",
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_topn_station_by_account",
"aliases": [
{
"text": "哪些站点用户最多？",
"tone": "口语"
},
{
"text": "按站点统计账户分布",
"tone": "中性"
},
{
"text": "站点账户数Top20排名",
"tone": "专业"
}
],
"keywords": [
"站点",
"账户数",
"TopN",
"排行",
"统计",
"station",
"account_id",
"去重",
"高负载",
"维度分析",
"by_dimension",
"aggregate",
"topn"
],
"intent_tags": [
"topn",
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_topn_district_by_account",
"aliases": [
{
"text": "哪个区用户最多？",
"tone": "口语"
},
{
"text": "按行政区统计账户数量",
"tone": "中性"
},
{
"text": "行政区账户数全量排名",
"tone": "专业"
}
],
"keywords": [
"行政区",
"账户数",
"TopN",
"排行",
"统计",
"district",
"account_id",
"去重",
"区域对比",
"维度分析",
"by_dimension",
"aggregate",
"topn"
],
"intent_tags": [
"topn",
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_share_of_meter_type",
"aliases": [
{
"text": "各类水表占多少比例？",
"tone": "口语"
},
{
"text": "水表类型占比分析",
"tone": "中性"
},
{
"text": "水表类型占比分布",
"tone": "专业"
}
],
"keywords": [
"水表类型",
"占比",
"比例",
"meter_type",
"account_id",
"去重",
"分布",
"主流类型",
"技术选型",
"ratio",
"aggregate",
"by_dimension"
],
"intent_tags": [
"ratio",
"aggregate",
"by_dimension"
]
},
{
"id": "snpt_sample_account_service_point",
"aliases": [
{
"text": "随机看10条账户信息",
"tone": "口语"
},
{
"text": "抽样账户与服务点明细",
"tone": "中性"
},
{
"text": "账户-服务点随机抽样验证",
"tone": "专业"
}
],
"keywords": [
"抽样",
"随机",
"样本",
"account_id",
"service_point_id",
"数据质量",
"验证",
"唯一性",
"格式检查",
"sample",
"quality"
],
"intent_tags": [
"sample",
"quality"
]
},
{
"id": "snpt_filter_meter_status_valid",
"aliases": [
{
"text": "只取有效的水表记录",
"tone": "口语"
},
{
"text": "筛选有效水表记录",
"tone": "中性"
},
{
"text": "水表状态有效性过滤",
"tone": "专业"
}
],
"keywords": [
"有效",
"过滤",
"筛选",
"meter_status",
"质量检查",
"断言",
"清洗",
"filter",
"quality"
],
"intent_tags": [
"filter",
"quality"
]
},
{
"id": "snpt_filter_meter_diameter_20mm",
"aliases": [
{
"text": "找出所有20mm水表用户",
"tone": "口语"
},
{
"text": "筛选20mm水表记录",
"tone": "中性"
},
{
"text": "20mm口径水表子集提取",
"tone": "专业"
}
],
"keywords": [
"20mm",
"水表直径",
"过滤",
"筛选",
"meter_diameter",
"子集",
"分析",
"住宅用水",
"规格",
"filter",
"by_dimension"
],
"intent_tags": [
"filter",
"by_dimension"
]
}
]

227
demo/水务/水务-qwen3-next-80b-snippet.json Normal file
View File

@ -0,0 +1,227 @@
[
{
"id": "snpt_topn_supply_office_by_account",
"desc": "统计各供应办公室对应的账户数量，识别高占比管理所",
"type": "topn",
"title": "按供应办公室统计账户数",
"examples": [
"哪个供水管理所服务的用户最多？",
"列出前5个账户数最多的供应办公室"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 11
}
],
"dialect_sql": {
"mysql": "SELECT supply_office, COUNT(DISTINCT account_id) AS account_count\nFROM water_meter_info\nGROUP BY supply_office\nORDER BY account_count DESC\nLIMIT {{top_n}};"
},
"applicability": {
"constraints": {
"notes": [
"供应办公室仅11个唯一值，可安全展示全部；建议LIMIT 11避免冗余排序"
],
"fk_join_available": false,
"dim_cardinality_hint": 11
},
"time_column": "nullable",
"required_columns": [
"supply_office",
"account_id"
]
},
"business_caliber": "粒度=供应办公室，指标=去重账户数（account_id），仅统计水表信息表中有效账户，不关联外部表"
},
{
"id": "snpt_topn_station_by_account",
"desc": "统计各站点服务的账户数量，识别高负载站点",
"type": "topn",
"title": "按站点统计账户分布",
"examples": [
"哪些站点服务的用户最多？",
"TOP10用户最多的站点是哪些？"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 20
}
],
"dialect_sql": {
"mysql": "SELECT station, COUNT(DISTINCT account_id) AS account_count\nFROM water_meter_info\nGROUP BY station\nORDER BY account_count DESC\nLIMIT {{top_n}};"
},
"applicability": {
"constraints": {
"notes": [
"站点有36个唯一值，建议LIMIT<=20以避免结果过长；高基数维度可能影响查询性能"
],
"fk_join_available": false,
"dim_cardinality_hint": 36
},
"time_column": "nullable",
"required_columns": [
"station",
"account_id"
]
},
"business_caliber": "粒度=站点（station），指标=去重账户数（account_id），基于水表信息表直接聚合，不涉及时间维度"
},
{
"id": "snpt_topn_district_by_account",
"desc": "统计各行政区的账户数量，辅助区域资源分配分析",
"type": "topn",
"title": "按行政区统计账户分布",
"examples": [
"哪个区的用水账户最多？",
"列出所有行政区的账户数量排名"
],
"variables": [
{
"name": "top_n",
"type": "int",
"default": 13
}
],
"dialect_sql": {
"mysql": "SELECT district, COUNT(DISTINCT account_id) AS account_count\nFROM water_meter_info\nGROUP BY district\nORDER BY account_count DESC\nLIMIT {{top_n}};"
},
"applicability": {
"constraints": {
"notes": [
"行政区共13个，可完整展示；适合用于区域对比分析"
],
"fk_join_available": false,
"dim_cardinality_hint": 13
},
"time_column": "nullable",
"required_columns": [
"district",
"account_id"
]
},
"business_caliber": "粒度=行政区（district），指标=去重账户数（account_id），基于水表信息表聚合，反映各区域用户规模"
},
{
"id": "snpt_share_of_meter_type",
"desc": "计算各类水表类型在总账户中的占比，识别主流类型",
"type": "ratio",
"title": "水表类型占比分析",
"examples": [
"各类水表在用户中的占比是多少？",
"电磁式远传水表占总用户比例多少？"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT meter_type, \n COUNT(DISTINCT account_id) AS account_count,\n ROUND(COUNT(DISTINCT account_id) * 100.0 / SUM(COUNT(DISTINCT account_id)) OVER (), 2) AS percentage\nFROM water_meter_info\nGROUP BY meter_type\nORDER BY account_count DESC;"
},
"applicability": {
"constraints": {
"notes": [
"水表类型仅5种，适合计算占比；可直接展示全量分布"
],
"fk_join_available": false,
"dim_cardinality_hint": 5
},
"time_column": "nullable",
"required_columns": [
"meter_type",
"account_id"
]
},
"business_caliber": "粒度=水表类型（meter_type），指标=去重账户数占比，分母为全表去重账户总数，反映技术选型分布"
},
{
"id": "snpt_sample_account_service_point",
"desc": "随机抽取部分账户与服务点ID的原始记录，用于数据质量核查",
"type": "sample",
"title": "抽样账户与服务点明细",
"examples": [
"随机查看10条账户与服务点的详细信息",
"抽样检查水表信息是否符合预期格式"
],
"variables": [
{
"name": "sample_size",
"type": "int",
"default": 10
}
],
"dialect_sql": {
"mysql": "SELECT account_id, service_point_id, supply_office, station, district, meter_diameter, meter_type, meter_subtype, installation_position\nFROM water_meter_info\nORDER BY RAND()\nLIMIT {{sample_size}};"
},
"applicability": {
"constraints": {
"notes": [
"主键组合为account_id+service_point_id，适合抽样验证唯一性；建议样本量≤100"
],
"fk_join_available": false,
"dim_cardinality_hint": null
},
"time_column": "nullable",
"required_columns": [
"account_id",
"service_point_id"
]
},
"business_caliber": "粒度=单条水表记录，抽取样本用于验证account_id与service_point_id的组合唯一性及维度字段完整性"
},
{
"id": "snpt_filter_meter_status_valid",
"desc": "过滤出水表状态为'有效'的记录，用于后续分析",
"type": "quality",
"title": "筛选有效水表记录",
"examples": [
"只取状态为有效的水表记录",
"确认所有水表是否均为有效状态"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT *\nFROM water_meter_info\nWHERE meter_status = '有效';"
},
"applicability": {
"constraints": {
"notes": [
"meter_status仅存在'有效'值，此条件恒成立；可用于数据清洗流程的显式过滤"
],
"fk_join_available": false,
"dim_cardinality_hint": 1
},
"time_column": "nullable",
"required_columns": [
"meter_status"
]
},
"business_caliber": "仅保留水表状态为'有效'的记录，因全表均为有效值，此过滤为冗余但可作为数据质量校验的显式断言"
},
{
"id": "snpt_filter_meter_diameter_20mm",
"desc": "筛选水表直径为20mm的记录，用于特定口径设备分析",
"type": "quality",
"title": "筛选20mm水表记录",
"examples": [
"找出所有使用20mm水表的用户",
"20mm水表分布在哪些站点？"
],
"variables": [],
"dialect_sql": {
"mysql": "SELECT *\nFROM water_meter_info\nWHERE meter_diameter = '20mm';"
},
"applicability": {
"constraints": {
"notes": [
"水表直径共8种枚举值，20mm为常见规格；可作为子集分析的起点"
],
"fk_join_available": false,
"dim_cardinality_hint": 8
},
"time_column": "nullable",
"required_columns": [
"meter_diameter"
]
},
"business_caliber": "粒度=单条水表记录，筛选条件为meter_diameter='20mm'，用于分析标准住宅用水表的分布特征"
}
]

21
file/ecommerce_orders.sql Normal file
View File

@ -0,0 +1,21 @@
CREATE TABLE `ecommerce_orders` (
`order_id` char(36) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'UUID from CSV',
`customer_id` int NOT NULL,
`product_id` int NOT NULL,
`category` varchar(64) COLLATE utf8mb4_unicode_ci NOT NULL,
`price` decimal(10,2) NOT NULL,
`quantity` int NOT NULL,
`order_date` datetime(6) NOT NULL,
`shipping_date` datetime(6) NOT NULL,
`delivery_status` varchar(32) COLLATE utf8mb4_unicode_ci NOT NULL,
`payment_method` varchar(32) COLLATE utf8mb4_unicode_ci NOT NULL,
`device_type` varchar(32) COLLATE utf8mb4_unicode_ci NOT NULL,
`channel` varchar(32) COLLATE utf8mb4_unicode_ci NOT NULL,
`shipping_address` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL,
`billing_address` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL,
`customer_segment` varchar(32) COLLATE utf8mb4_unicode_ci NOT NULL,
PRIMARY KEY (`order_id`),
KEY `idx_customer` (`customer_id`),
KEY `idx_product` (`product_id`),
KEY `idx_order_date` (`order_date`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci;