227 lines
8.9 KiB
JSON
227 lines
8.9 KiB
JSON
[
|
||
{
|
||
"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"
|
||
}
|
||
] |