首页 > 技术文章 > Elasticsearch 之(34)基于term vector深入探查数据的情况

wuzhiwei549 2018-05-31 13:22 原文

1、term vector介绍

获取document中的某个field内的各个term的统计信息

term information: term frequency in the field, term positions, start and end offsets, term payloads

term statistics: 设置term_statistics=true; total term frequency, 一个term在所有document中出现的频率; document frequency,有多少document包含这个term
field statistics: document count,有多少document包含这个field; sum of document frequency,一个field中所有term的df之和; sum of total term frequency,一个field中的所有term的tf之和

GET /twitter/tweet/1/_termvectors
GET /twitter/tweet/1/_termvectors?fields=text

term statistics和field statistics并不精准,不会被考虑有的doc可能被删除了
其实很少用,用的时候,一般来说,就是你需要对一些数据做探查的时候。比如说,你想要看到某个term,某个词条,大话西游,这个词条,在多少个document中出现了。或者说某个field,film_desc,电影的说明信息,有多少个doc包含了这个说明信息。

2、index-time term vector实验

掌握,如何采集term vector信息,然后如何看懂term vector信息,你能掌握利用term vector进行数据探查term vector,涉及了很多的term和field相关的统计信息,有两种方式可以采集到这个统计信息
(1)index-time,你在mapping里配置一下,然后建立索引的时候,就直接给你生成这些term和field的统计信息了
(2)query-time,你之前没有生成过任何的Term vector信息,然后在查看term vector的时候,直接就可以看到了,会on the fly,现场计算出各种统计信息,然后返回给你
设置mapping
PUT /my_index
{
  "mappings": {
    "my_type": {
      "properties": {
        "text": {
            "type": "text",
            "term_vector": "with_positions_offsets_payloads",
            "store" : true,
            "analyzer" : "fulltext_analyzer"
         },
         "fullname": {
            "type": "text",
            "analyzer" : "fulltext_analyzer"
        }
      }
    }
  },
  "settings" : {
    "index" : {
      "number_of_shards" : 1,
      "number_of_replicas" : 0
    },
    "analysis": {
      "analyzer": {
        "fulltext_analyzer": {
          "type": "custom",
          "tokenizer": "whitespace",
          "filter": [
            "lowercase",
            "type_as_payload"
          ]
        }
      }
    }
  }
}
初始化数据
PUT /my_index/my_type/1
{
  "fullname" : "Leo Li",
  "text" : "hello test test test "
}

PUT /my_index/my_type/2
{
  "fullname" : "Leo Li",
  "text" : "other hello test ..."
}
_termvectors探查数据
GET /my_index/my_type/1/_termvectors
{
  "fields" : ["text"],
  "offsets" : true,
  "payloads" : true,
  "positions" : true,
  "term_statistics" : true,
  "field_statistics" : true
}
{
  "_index": "my_index",
  "_type": "my_type",
  "_id": "1",
  "_version": 1,
  "found": true,
  "took": 10,
  "term_vectors": {
    "text": {
      "field_statistics": {
        "sum_doc_freq": 6,  //sum of document  所有field中的所有term的tf(term frequency)之和
        "doc_count": 2,     //查询document的数量
        "sum_ttf": 8       //sum of total term frequency,所有field中的所有term的tf(term frequency)之和
      },
      "terms": {
        "hello": {
          "doc_freq": 2,   //document frequency,有多少document包含这个term
          "ttf": 2,		   // total term frequency, 一个term在所有document中出现的频率
          "term_freq": 1,  // term frequency, 一个term在当前document中出现的频率
          "tokens": [      //hello这个term的信息
            {
              "position": 0,      //hello在field中的位置
              "start_offset": 0,  //hello的首字母在field中的位置
              "end_offset": 5,    //hello的最后一个字母在field中的位置
              "payload": "d29yZA=="
            }
          ]
        },
        "test": {
          "doc_freq": 2,
          "ttf": 4,
          "term_freq": 3,
          "tokens": [
            {
              "position": 1,
              "start_offset": 6,
              "end_offset": 10,
              "payload": "d29yZA=="
            },
            {
              "position": 2,
              "start_offset": 11,
              "end_offset": 15,
              "payload": "d29yZA=="
            },
            {
              "position": 3,
              "start_offset": 16,
              "end_offset": 20,
              "payload": "d29yZA=="
            }
          ]
        }
      }
    }
  }
}

3、query-time term vector实验

一般来说,如果条件允许,你就用query time的term vector就可以了,你要探查什么数据,现场去探查一下就好了
GET /my_index/my_type/1/_termvectors
{
  "fields" : ["fullname"],
  "offsets" : true,
  "positions" : true,
  "term_statistics" : true,
  "field_statistics" : true
}

4、手动指定doc的term vector

GET /my_index/my_type/_termvectors
{
  "doc" : {
    "fullname" : "Leo Li",
    "text" : "hello test test test"
  },
  "fields" : ["text"],
  "offsets" : true,
  "payloads" : true,
  "positions" : true,
  "term_statistics" : true,
  "field_statistics" : true
}

手动指定一个doc,实际上不是要指定doc,而是要指定你想要安插的词条,hello test,那么就可以放在一个field中
将这些term分词,然后对每个term,都去计算它在现有的所有doc中的一些统计信息
这个挺有用的,可以让你手动指定要探查的term的数据情况,你就可以指定探查“大话西游”这个词条的统计信息

5、手动指定analyzer来生成term vector
GET /my_index/my_type/_termvectors
{
  "doc" : {
    "fullname" : "Leo Li",
    "text" : "hello test test test"
  },
  "fields" : ["text"],
  "offsets" : true,
  "payloads" : true,
  "positions" : true,
  "term_statistics" : true,
  "field_statistics" : true,
  "per_field_analyzer" : {
    "text": "standard"
  }
}

6、terms filter
max_num_terms 最大的词条数目
min_term_freq 最小的词频,比如忽略那些在字段中出现次数小于一定值的词条。
max_term_freq 最大的词频
min_doc_freq 最小的文档频率,比如忽略那些在文档中出现次数小于一定的值的词条
max_doc_freq 最大的文档频率
min_word_length 忽略的词的最小长度
max_word_length 忽略的词的最大长度

GET /my_index/my_type/_termvectors
{
  "doc" : {
    "fullname" : "Leo Li",
    "text" : "hello test test test"
  },
  "fields" : ["text"],
  "offsets" : true,
  "payloads" : true,
  "positions" : true,
  "term_statistics" : true,
  "field_statistics" : true,
  "filter" : {
      "max_num_terms" : 3,
      "min_term_freq" : 1,
      "min_doc_freq" : 1
    }
}
这个就是说,根据term统计信息,过滤出你想要看到的term vector统计结果
也挺有用的,比如你探查数据把,可以过滤掉一些出现频率过低的term,就不考虑了

7、multi term vector手动指定查询多个doc的term vector
GET _mtermvectors
{
   "docs": [
      {
         "_index": "my_index",
         "_type": "my_type",
         "_id": "2",
         "term_statistics": true
      },
      {
         "_index": "my_index",
         "_type": "my_type",
         "_id": "1",
         "fields": [
            "text"
         ]
      }
   ]
}
GET /my_index/_mtermvectors
{
   "docs": [
      {
         "_type": "test",
         "_id": "2",
         "fields": [
            "text"
         ],
         "term_statistics": true
      },
      {
         "_type": "test",
         "_id": "1"
      }
   ]
}
GET /my_index/my_type/_mtermvectors
{
   "docs": [
      {
         "_id": "2",
         "fields": [
            "text"
         ],
         "term_statistics": true
      },
      {
         "_id": "1"
      }
   ]
}
GET /_mtermvectors
{
   "docs": [
      {
         "_index": "my_index",
         "_type": "my_type",
         "doc" : {
            "fullname" : "Leo Li",
            "text" : "hello test test test"
         }
      },
      {
         "_index": "my_index",
         "_type": "my_type",
         "doc" : {
           "fullname" : "Leo Li",
           "text" : "other hello test ..."
         }
      }
   ]
}




推荐阅读