什么是全文檢索

和術(shù)語級別查詢（Term-Level Queries）不同，全文檢索查詢（Full Text Queries）旨在基于相關(guān)性搜索和匹配文本數(shù)據(jù)。這些查詢會對輸入的文本進(jìn)行分析，將其拆分為詞項(xiàng)（單個單詞），并執(zhí)行諸如分詞、詞干處理和標(biāo)準(zhǔn)化等操作。

全文檢索的關(guān)鍵特點(diǎn)：

• 對輸入的文本進(jìn)行分析，并根據(jù)分析后的詞項(xiàng)進(jìn)行搜索和匹配。全文檢索查詢會對輸入的文本進(jìn)行分析，將其拆分為詞項(xiàng)，并基于這些詞項(xiàng)進(jìn)行搜索和匹配操作。
• 以相關(guān)性為基礎(chǔ)進(jìn)行搜索和匹配。全文檢索查詢使用相關(guān)性算法來確定文檔與查詢的匹配程度，并按照相關(guān)性進(jìn)行排序。相關(guān)性可以基于詞項(xiàng)的頻率、權(quán)重和其他因素來計(jì)算。
• 全文檢索查詢適用于包含自由文本數(shù)據(jù)的字段，例如文檔的內(nèi)容、文章的正文或產(chǎn)品描述等。

一、數(shù)據(jù)準(zhǔn)備

PUT full_index
{
  "settings": {
    "number_of_replicas": 1,
    "number_of_shards": 1
  },
  "mappings": {
    "properties": {
      "name": {
        "type": "text"
      },
      "age": {
        "type": "long"
      },
      "description" : {
          "type" : "text",
          "analyzer": "ik_max_word",
          "fields" : {
            "keyword" : {
              "type" : "keyword",
              "ignore_above" : 256
            }
          }
        }
    }
  }
}

測試數(shù)據(jù)如下:
{name=張三, description=北京故宮圓明園, age=11}
{name=王五, description=南京總統(tǒng)府, age=15}
{name=李四, description=北京市天安門廣場, age=18}
{name=富貴, description=南京市中山陵, age=22}
{name=來福, description=山東濟(jì)南趵突泉, age=8}
{name=憨憨, description=安徽黃山九華山, age=27}
{name=小七, description=上海東方明珠, age=31}

二、match query

匹配查詢: match在匹配時會對所查找的關(guān)鍵詞進(jìn)行分詞，然后按分詞匹配查找。

match支持以下參數(shù)：

• query : 指定匹配的值
• operator : 匹配條件類型
• and : 條件分詞后都要匹配
• or : 條件分詞后有一個匹配即可(默認(rèn))
• minmum_should_match : 最低匹配度，即條件在倒排索引中最低的匹配度

DSL: 索引description字段包含 “南京總統(tǒng)府” 的數(shù)據(jù)

GET  full_index/_search
{
  "query": {
    "match": {
      "description": "南京總統(tǒng)府"
    }
  }
}

返回?cái)?shù)據(jù)如下:
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 2,
      "relation" : "eq"
    },
    "max_score" : 1.2667978,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 1.2667978,
        "_source" : {
          "name" : "王五",
          "age" : 15,
          "description" : "南京總統(tǒng)府"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "4",
        "_score" : 1.0751815,
        "_source" : {
          "name" : "富貴",
          "age" : 22,
          "description" : "南京市中山陵"
        }
      }
    ]
  }
}

springboot實(shí)現(xiàn):

    private final static Logger LOGGER = LoggerFactory.getLogger(FullTextQuery.class);

    private static final String INDEX_NAME = "full_index";

    @Resource
    private RestHighLevelClient client;
    
    @RequestMapping(value = "/match_query", method = RequestMethod.GET)
    @ApiOperation(value = "DSL - match_query")
    public void match_query() throws Exception {
        // 定義請求對象
        SearchRequest searchRequest = new SearchRequest(INDEX_NAME);
        // 查詢所有
        searchRequest.source(new SearchSourceBuilder().query(QueryBuilders.matchQuery("description","南京總統(tǒng)府")));
        // 打印返回?cái)?shù)據(jù)
        printLog(client.search(searchRequest, RequestOptions.DEFAULT));
    }

    private void printLog(SearchResponse searchResponse) {
        SearchHits hits = searchResponse.getHits();
        System.out.println("返回hits數(shù)組長度:" + hits.getHits().length);
        for (SearchHit hit: hits.getHits()) {
            System.out.println(hit.getSourceAsMap().toString());
        }
    }
    
返回結(jié)果如下:
返回hits數(shù)組長度:2
{name=王五, description=南京總統(tǒng)府, age=15}
{name=富貴, description=南京市中山陵, age=22}

分析： 此時可以發(fā)現(xiàn)當(dāng)搜索 “南京總統(tǒng)府” 時，返回了兩條數(shù)據(jù)，那么為什么 “南京市中山陵” 也被搜索到了呢？
原因就是全文檢索會拆分搜索的此項(xiàng)，因?yàn)樵趧?chuàng)建索引的時候指定了 description 字段的分詞方式是 “ik_max_word” ，而該分詞類型會將 “南京總統(tǒng)府” 拆分成如下詞類去搜索倒排索引:

POST _analyze
{
  "analyzer": "ik_max_word",
  "text": ["南京總統(tǒng)府"]
}

{
  "tokens" : [
    {
      "token" : "南京",
      "start_offset" : 0,
      "end_offset" : 2,
      "type" : "CN_WORD",
      "position" : 0
    },
    {
      "token" : "總統(tǒng)府",
      "start_offset" : 2,
      "end_offset" : 5,
      "type" : "CN_WORD",
      "position" : 1
    },
    {
      "token" : "總統(tǒng)",
      "start_offset" : 2,
      "end_offset" : 4,
      "type" : "CN_WORD",
      "position" : 2
    },
    {
      "token" : "府",
      "start_offset" : 4,
      "end_offset" : 5,
      "type" : "CN_CHAR",
      "position" : 3
    }
  ]
}

其中就有"南京"這個詞項(xiàng)，所以用 “南京總統(tǒng)府” 去搜索是可以搜到 “南京中山陵” 的數(shù)據(jù)，那么match_query的operator也不用多說，就是滿足所有拆分的詞項(xiàng)

比如此時我們再插入一條數(shù)據(jù):
POST /full_index/_bulk
{"index":{"_id":8}}
{"name":"張三","age":11,"description":"南京總統(tǒng)"}

當(dāng)我們搜索:"南京總統(tǒng)",可以搜到兩條數(shù)據(jù)
GET  full_index/_search
{
  "query": {
    "match": {
      "description": {
        "query": "南京總統(tǒng)",
        "operator": "and"
      }
    }
  }
}
數(shù)據(jù)如下:
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 2,
      "relation" : "eq"
    },
    "max_score" : 2.898355,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "8",
        "_score" : 2.898355,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "南京總統(tǒng)"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 2.35562,
        "_source" : {
          "name" : "王五",
          "age" : 15,
          "description" : "南京總統(tǒng)府"
        }
      }
    ]
  }
}

但是當(dāng)搜索:"南京總統(tǒng)府"時，只能搜索到一條數(shù)據(jù),就是因?yàn)榉衷~時，有一個詞項(xiàng)"府",在其中一條數(shù)據(jù)中不存在

三、multi_match query

多字段查詢:可以根據(jù)字段類型，決定是否使用分詞查詢，得分最高的在前面
注意：字段類型分詞,將查詢條件分詞之后進(jìn)行查詢，如果該字段不分詞就會將查詢條件作為整體進(jìn)行查詢。

DSL: 查詢 “name” 或者 “description” 這兩個字段中出現(xiàn) “北京王五” 詞匯的數(shù)據(jù)

GET  full_index/_search
{
  "query": {
    "multi_match": {
      "query": "北京王五",
      "fields": ["name","description"]
    }
  }
}

返回結(jié)果如下:
{
  "took" : 1,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 3,
      "relation" : "eq"
    },
    "max_score" : 3.583519,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 3.583519,
        "_source" : {
          "name" : "王五",
          "age" : 15,
          "description" : "南京總統(tǒng)府"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 1.4959542,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "北京故宮圓明園"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "3",
        "_score" : 0.98645234,
        "_source" : {
          "name" : "李四",
          "age" : 18,
          "description" : "北京市天安門廣場"
        }
      }
    ]
  }
}

springboot實(shí)現(xiàn):

    @RequestMapping(value = "/multi_match", method = RequestMethod.GET)
    @ApiOperation(value = "DSL - multi_match")
    public void multi_match() throws Exception {
        // 定義請求對象
        SearchRequest searchRequest = new SearchRequest(INDEX_NAME);
        // 查詢所有
        searchRequest.source(new SearchSourceBuilder().query(
                QueryBuilders.multiMatchQuery("北京王五", new String[]{"name","description"})));
        // 打印返回?cái)?shù)據(jù)
        printLog(client.search(searchRequest, RequestOptions.DEFAULT));
    }

查詢結(jié)果如下:
返回hits數(shù)組長度:3
{name=王五, description=南京總統(tǒng)府, age=15}
{name=張三, description=北京故宮圓明園, age=11}
{name=李四, description=北京市天安門廣場, age=18}

前面也強(qiáng)調(diào)到 字段類型分詞,將查詢條件分詞之后進(jìn)行查詢，如果該字段不分詞就會將查詢條件作為整體進(jìn)行查詢
那么我們來測試一下,比如當(dāng)不對 “description” 字段分詞時查詢

GET  full_index/_search
{
  "query": {
    "multi_match": {
      "query": "北京王五",
      "fields": ["name","description.keyword"]
    }
  }
}
返回結(jié)果如下:
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 3.583519,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "2",
        "_score" : 3.583519,
        "_source" : {
          "name" : "王五",
          "age" : 15,
          "description" : "南京總統(tǒng)府"
        }
      }
    ]
  }
}

可以看到，當(dāng)使用 “description.keyword” 也就是不對 “description” 進(jìn)行分詞時，只返回了一條數(shù)據(jù)，該條數(shù)據(jù)只有 “name” 字段為 “王五” 滿足了查詢條件分詞匹配后的結(jié)果。

四、match_phrase query

短語搜索(match phrase)會對搜索文本進(jìn)行文本分析，然后到索引中尋找搜索的每個分詞并要求分詞相鄰，你可以通過調(diào)整slop參數(shù)設(shè)置分詞出現(xiàn)的最大間隔距離。match_phrase 會將檢索關(guān)鍵詞分詞。

DSL: 搜索 "description " 字段有 “北京故宮” 的數(shù)據(jù)

GET  full_index/_search
{
  "query": {
    "match_phrase": {
      "description": {
        "query": "北京故宮"
      }
    }
  }
}

返回?cái)?shù)據(jù)如下:
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 3.5884824,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 3.5884824,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "北京故宮圓明園"
        }
      }
    ]
  }
}

springboot實(shí)現(xiàn):

    @RequestMapping(value = "/match_phrase", method = RequestMethod.GET)
    @ApiOperation(value = "DSL - match_phrase")
    public void match_phrase() throws Exception {
        // 定義請求對象
        SearchRequest searchRequest = new SearchRequest(INDEX_NAME);
        // 查詢所有
        searchRequest.source(new SearchSourceBuilder().query(
                QueryBuilders.matchPhraseQuery("description","北京故宮")));
        // 打印返回?cái)?shù)據(jù)
        printLog(client.search(searchRequest, RequestOptions.DEFAULT));
    }

返回?cái)?shù)據(jù)如下:
返回hits數(shù)組長度:1
{name=張三, description=北京故宮圓明園, age=11}

思考： 搜索 "description " 字段有 “北京故宮” 的數(shù)據(jù)有返回，那么搜索 “北京圓明園” ，為什么沒數(shù)據(jù)返回？

GET  full_index/_search
{
  "query": {
    "match_phrase": {
      "description": {
        "query": "北京圓明園"
      }
    }
  }
}
返回?cái)?shù)據(jù)如下:
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 0,
      "relation" : "eq"
    },
    "max_score" : null,
    "hits" : [ ]
  }
}

原因分析: 先查看 “北京故宮圓明園” 的分詞結(jié)果，如下:

POST _analyze
{
  "analyzer": "ik_max_word",
  "text": ["北京故宮圓明園"]
}

{
  "tokens" : [
    {
      "token" : "北京",
      "start_offset" : 0,
      "end_offset" : 2,
      "type" : "CN_WORD",
      "position" : 0
    },
    {
      "token" : "故宮",
      "start_offset" : 2,
      "end_offset" : 4,
      "type" : "CN_WORD",
      "position" : 1
    },
    {
      "token" : "圓明園",
      "start_offset" : 4,
      "end_offset" : 7,
      "type" : "CN_WORD",
      "position" : 2
    }
  ]
}

可以發(fā)現(xiàn) “北京” 和 “圓明園” 并不是相鄰的詞條,他們之間相差了一個詞條，所以這時候就需要用到 “slop” , slop參數(shù)告訴match_phrase查詢詞條能夠相隔多遠(yuǎn)時仍然將文檔視為匹配

GET  full_index/_search
{
  "query": {
    "match_phrase": {
      "description": {
        "query": "北京圓明園",
        "slop": 1
      }
    }
  }
}
返回結(jié)果如下:
{
  "took" : 1,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 2.4425511,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 2.4425511,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "北京故宮圓明園"
        }
      }
    ]
  }
}

五、query_string query

允許我們在單個查詢字符串中指定AND | OR | NOT條件，同時也和 multi_match query 一樣，支持多字段搜索。和match類似，但是match需要指定字段名，query_string是在所有字段中搜索，范圍更廣泛。
注意: 查詢字段分詞就將查詢條件分詞查詢，查詢字段不分詞將查詢條件不分詞查詢

DSL: 搜索當(dāng)前索引所有字段中含有 “北京故宮” 的文檔

GET  full_index/_search
{
  "query": {
    "query_string": {
      "query": "安徽張三"
    }
  }
}

返回?cái)?shù)據(jù)如下:
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 3,
      "relation" : "eq"
    },
    "max_score" : 2.5618675,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 2.5618675,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "北京故宮圓明園"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "8",
        "_score" : 2.5618675,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "南京總統(tǒng)"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "6",
        "_score" : 1.7342355,
        "_source" : {
          "name" : "憨憨",
          "age" : 27,
          "description" : "安徽黃山九華山"
        }
      }
    ]
  }
}

springboot實(shí)現(xiàn)：

    @RequestMapping(value = "/query_string", method = RequestMethod.GET)
    @ApiOperation(value = "DSL - query_string")
    public void query_string() throws Exception {
        // 定義請求對象
        SearchRequest searchRequest = new SearchRequest(INDEX_NAME);
        // 查詢所有
        searchRequest.source(new SearchSourceBuilder().query(
                QueryBuilders.queryStringQuery("安徽張三")));
        // 打印返回?cái)?shù)據(jù)
        printLog(client.search(searchRequest, RequestOptions.DEFAULT));
    }

返回hits數(shù)組長度:3
{name=張三, description=北京故宮圓明園, age=11}
{name=張三, description=南京總統(tǒng), age=11}
{name=憨憨, description=安徽黃山九華山, age=27}

指定字段查詢: “description” 字段中含有 “安徽張三” 的文檔

GET  full_index/_search
{
  "query": {
    "query_string": {
      "query": "安徽張三",
      "fields": ["description"]
    }
  }
}

返回?cái)?shù)據(jù)如下：
{
  "took" : 0,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 1.7342355,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "6",
        "_score" : 1.7342355,
        "_source" : {
          "name" : "憨憨",
          "age" : 27,
          "description" : "安徽黃山九華山"
        }
      }
    ]
  }
}

指定多個字段查詢 ： 查詢 “安徽” “憨憨” 同時滿足

GET  full_index/_search
{
  "query": {
    "query_string": {
      "query": "安徽 AND 憨憨",
      "fields": ["description","name"]
    }
  }
}

返回:
{
  "took" : 1,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 1,
      "relation" : "eq"
    },
    "max_score" : 6.6615744,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "6",
        "_score" : 6.6615744,
        "_source" : {
          "name" : "憨憨",
          "age" : 27,
          "description" : "安徽黃山九華山"
        }
      }
    ]
  }
}

GET  full_index/_search
{
  "query": {
    "query_string": {
      "query": "(安徽 AND 憨憨)OR 張三",
      "fields": ["description","name"]
    }
  }
}
返回?cái)?shù)據(jù)如下:
{
  "took" : 3,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 3,
      "relation" : "eq"
    },
    "max_score" : 6.6615744,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "6",
        "_score" : 6.6615744,
        "_source" : {
          "name" : "憨憨",
          "age" : 27,
          "description" : "安徽黃山九華山"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 2.5618675,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "北京故宮圓明園"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "8",
        "_score" : 2.5618675,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "南京總統(tǒng)"
        }
      }
    ]
  }
}

query_string query 這種查詢方式類似于 match query匹配查詢 結(jié)合 multi_match query 多字段查詢 一起使用。

六、simple_query_string

類似Query String，但是會忽略錯誤的語法,同時只支持部分查詢語法，不支持AND OR NOT，會當(dāng)作字符串處理。支持部分邏輯：文章來源地址http://www.zghlxwxcb.cn/news/detail-814195.html

• “+” 替代 “AND”
• “|” 替代 “OR”
• “-” 替代 “NOT”

GET full_index/_search
{
  "query": {
    "simple_query_string": {
      "query": "(安徽 + 憨憨) | 張三",
      "fields": ["description","name"]
    }
  }
}

返回結(jié)果如下:
{
  "took" : 41,
  "timed_out" : false,
  "_shards" : {
    "total" : 1,
    "successful" : 1,
    "skipped" : 0,
    "failed" : 0
  },
  "hits" : {
    "total" : {
      "value" : 3,
      "relation" : "eq"
    },
    "max_score" : 6.6615744,
    "hits" : [
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "6",
        "_score" : 6.6615744,
        "_source" : {
          "name" : "憨憨",
          "age" : 27,
          "description" : "安徽黃山九華山"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "1",
        "_score" : 2.5618675,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "北京故宮圓明園"
        }
      },
      {
        "_index" : "full_index",
        "_type" : "_doc",
        "_id" : "8",
        "_score" : 2.5618675,
        "_source" : {
          "name" : "張三",
          "age" : 11,
          "description" : "南京總統(tǒng)"
        }
      }
    ]
  }
}

到了這里，關(guān)于【ElasticSearch-基礎(chǔ)篇】ES高級查詢Query DSL全文檢索的文章就介紹完了。如果您還想了解更多內(nèi)容，請?jiān)谟疑辖撬阉鱐OY模板網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關(guān)文章，希望大家以后多多支持TOY模板網(wǎng)！