elasticsearch 桶(Bucket)聚合

1、Terms Aggregation
对词条Term字段的值进行group聚合，得到每个Term字段的值的个数

{
   "size" : 0,
   "aggs": {
      "colors": {
         "terms": {
            "field": "securityId"
         }
      }
   }
}
1
2
3
4
5
6
7
8
9
10

2、Histogram Aggregation直方图聚合
根据指定的时间间隔构造存储桶。 属于每个间隔的值将形成一个间隔存储桶。
interval参数指定每统计间隔是多少。

{
	"size":0,
  "aggs": {
      "colors": {
         "histogram": {
            "field": "priceNow",
            "interval": 100
         }
      }
   }
}
1
2
3
4
5
6
7
8
9
10
11

3、日期直方图聚合(Date histogram aggregation)

似于正常的直方图,区别在于，可以使用日期/时间表达式指定间隔。

{
  "size": 0,
  "aggs": {
    "birthdays": {
      "date_histogram": {
        "field": "updateTime",
        "interval": "year"
      }
    }
  }
}
1
2
3
4
5
6
7
8
9
10
11

4、 Range Aggregation范围值区间聚合
对指定字段的值，进行区间访问统计，得到各区间的count数

{
  "size":0,
  "query":{
    "bool":{
      "must":[
        {
          "range":{
            "tradeDate":{
               "from":20201216,
               "to":20201216,
               "include_lower":true,
               "include_upper":true
            }
          }
        }
      ]
    }
  },
  "aggs": {
    "goal_ranges": {
      "range": {
        "field": "priceNow",
        "ranges": [
          {
            "to": 20
          },
          {
            "from": 20,
            "to": 100
          },
          {
            "from": 100
          }
        ]
      }
    }
  }
}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38

5、聚合筛选，类似SQL里面的having

逻辑：指定日期范围内，按股票分组，求平均价值大于10，且在指定时间范围内只出现过1次的股票。

{
  "size":0,
  "query":{
    "bool":{
      "must":[
        {
          "range":{
            "tradeDate":{
               "from":20201213,
               "to":20201216,
               "include_lower":true,
               "include_upper":true
            }
          }
        }
      ]
    }
  },
  "aggs": {
    "groupUserId": {
      "terms": {
        "field": "securityId",
        "size":2000,
        "order" : { "_key" : "asc" }
      },
      "aggs": {
        "avgAmount": {
          "avg": {
            "field": "priceNow"
          }
        },
        "having": {
          "bucket_selector": {
            "buckets_path": {
              "orderCount": "_count",
              "avgAmount": "avgAmount"
            },
            "script": {
              "source": "params.avgAmount >= 10 && params.orderCount ==1 "
            }
          }
        }
      }
    }
  }
}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

6、邻近矩阵聚合(Adjacency matrix)

adjacency_matrix类型的聚合可以返回类似于集合的操作，例如A,B,C三个筛选条件，聚合请求就会返回符合A的，符合A&B的，符合B的，符合B&C的，符合C的，符合A&C的不同集合的数据数目。

{
  "size":0,
  "query":{
    "bool":{
      "must":[
        {
          "range":{
            "tradeDate":{
               "from":20201216,
               "to":20201216,
               "include_lower":true,
               "include_upper":true
            }
          }
        }
      ]
    }
  },
  "aggs" : {
    "interactions" : {
      "adjacency_matrix" : {
        "filters" : {
          "grpA" : { "terms" : { "signalType" : ["fundDepthSignal", "nxtjBsSignal"] }},
          "grpB" : { "terms" : { "signalType" : ["nxtjBsSignal", "zhcmBsSignal"] }},
          "grpC" : { "terms" : { "signalType" : ["fundDepthSignal", "bhlnEnergyBsSignal"] }}
        }
      }
    }
  }
}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30