我的索引中的 ElasticSearch 同义词用法/语法答案

【问题标题】：ElasticSearch Synonym usage / syntax in my index我的索引中的 ElasticSearch 同义词用法/语法
【发布时间】：2026-02-15 19:35:01
【问题描述】：

我是 ElasticSearch 的新手。我正在 ES 中尝试一个简单的家谱项目，并希望对名字和姓氏使用同义词。我有以下 ElasticSearch 索引设置，我希望在我的设置中添加两个同义词分析器，然后使用这两个在不同字段上进行搜索时同义词处理。我不知道如何将正确的条目添加到我的“设置”JSON 文件中

我的两个同义词文件分别称为 given_synonyms.txt 和 surname_synonyms.txt。我将不胜感激在获得此 json 文件的语法方面的帮助。

我已经为我想做的事情添加了（大写的）非常通用的伪代码，希望这样对读者更有意义。

    {
      "settings": {
        "index": {
          "number_of_shards": "128",
          "number_of_replicas": "0",
          "analysis": {
            "filter": {
              "dbl_metaphone": {
                "type":    "phonetic",
                "encoder": "double_metaphone",
                "max_code_len" : 5
              }
            },
            "analyzer": {
              "dbl_metaphone": {
                "tokenizer": "standard",
                "filter":    "dbl_metaphone"
              }
            }
    I ASSUME THE SYNONYM ENTRIES GO HERE FOR THE TWO SYNONYM FILES - BUT I DON'T KNOW HOW THAT SHOULD BE DONE
          }
        }
      },
      "mappings": {
        "test": {
          "_all": {
            "enabled": false
          },
          "_source": {
            "enabled": true
          },
          "properties": {
            "GivenNames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
                I WANT TO USE THE GIVEN_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
              }
            },
            "Surnames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
                I WANT TO USE THE SURNAME_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
              }
            },
            "FatherGivenNames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE GIVEN_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "FatherSurnames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE SURNAME_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "MotherGivenNames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE GIVEN_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "MotherSurnames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE SURNAME_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "SpouseGivenNames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE GIVEN_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "SpouseSurnames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE SURNAME_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "ChildrenGivenNames": {
              "type": "keyword",
              "index_options": "freqs",
              "store": "false",
              "similarity": "boolean",
              "norms": "false",
              "fields": {
                "phonetic": {
                  "type": "text",
                  "analyzer": "dbl_metaphone"
                }
              }
              I WANT TO USE THE GIVEN_SYNONYMS SEARCH ANALYZER HERE IN ADDITION TO THE DBL_METAPHONE
            },
            "BirthYears": {
              "type": "short"
            },
            "BirthLocations": {
              "type": "integer"
            },
            "DeathYears": {
              "type": "short"
            },
            "DeathLocations": {
              "type": "integer"
            },
            "MarriageLocations": {
              "type": "integer"
            },
            "MarriageYears": {
              "type": "integer"
            },
            "ResidenceLocations": {
              "type": "integer"
            }
          }
        }
      }
    }

【问题讨论】：

标签： elasticsearch synonym

【解决方案1】：

首先，让我们看看同义词分析器以及如何包含文件。从这里https://www.elastic.co/guide/en/elasticsearch/reference/current/analysis-synonym-tokenfilter.html 你可以看到如何配置它。你提到了两个文件。您最好定义一个包含您的同义词或同义词的文件。

{
  "settings": {
    "index": {
      "number_of_shards": "128",
      "number_of_replicas": "0",
      "analysis": {
        "filter": {
          "dbl_metaphone": {
            "type": "phonetic",
            "encoder": "double_metaphone",
            "max_code_len": 5
          },
          "given_synonyms": {
            "type": "synonym",
            "synonyms_path": "analysis/given_synonyms.txt"
          },
          "surname_synonyms": {
            "type": "synonym",
            "synonyms_path": "analysis/surname_synonyms.txt"
          }
        },
        "analyzer": {
          "dbl_metaphone": {
            "tokenizer": "standard",
            "filter": "dbl_metaphone"
          },
          "given_synonyms": {
            "tokenizer": "whitespace",
            "filter": "given_synonyms"
          },
          "surname_synonyms": {
            "tokenizer": "whitespace",
            "filter": "surname_synonyms"
          }
        }
      }
    }
  }
}

请记住，当您使用一个文件（或多个文件）作为同义词时，您需要确保每个 elasticsearch 节点都可以访问该文件。另一种方法是在设置部分指定同义词。如果同义词的数量不是很大，或者在使文件可供 elasticsearch 节点访问时存在问题，这可能会更好。您可以在文档中查看更多定义同义词的方法

{
  "settings": {
    "index": {
      "number_of_shards": "128",
      "number_of_replicas": "0",
      "analysis": {
        "filter": {
          "dbl_metaphone": {
            "type": "phonetic",
            "encoder": "double_metaphone",
            "max_code_len": 5
          },
          "given_synonyms": {
            "type": "synonym",
            "synonyms": [ ... YOUR SYNONYMS HERE ...]
          },
          "surname_synonyms": {
            "type": "synonym",
            "synonyms": [ ... YOUR SYNONYMS HERE ...]
          }
        },
        "analyzer": {
          "dbl_metaphone": {
            "tokenizer": "standard",
            "filter": "dbl_metaphone"
          },
          "given_synonyms": {
            "tokenizer": "standard",
            "filter": "given_synonyms"
          },
          "surname_synonyms": {
            "tokenizer": "standard",
            "filter": "surname_synonyms"
          }
        }
      }
    }
  }
}

为了达到最终的解决方案，你可以有这样的东西

{
  "settings": {
    "index": {
      "number_of_shards": "128",
      "number_of_replicas": "0",
      "analysis": {
        "filter": {
          "dbl_metaphone": {
            "type": "phonetic",
            "encoder": "double_metaphone",
            "max_code_len": 5
          },
          "given_synonyms": {
            "type": "synonym",
            "synonyms_path": "analysis/given_synonyms.txt"
          },
          "surname_synonyms": {
            "type": "synonym",
            "synonyms_path": "analysis/surname_synonyms.txt"
          }
        },
        "analyzer": {
          "dbl_metaphone": {
            "tokenizer": "standard",
            "filter": "dbl_metaphone"
          },
          "dbl_metaphone_given_synonym": {
            "tokenizer": "standard",
            "filter": [
              "given_synonyms",
              "dbl_metaphone"
            ]
          },
          "dbl_metaphone_surname_synonym": {
            "tokenizer": "standard",
            "filter": [
              "surname_synonyms",
              "dbl_metaphone"
            ]
          }
        }
      }
    }
  }
}

在此，您总共拥有三个分析器。其中两个正在组合两个过滤器（第一个的输出是第二个的输入，所以顺序很重要）。在elasticsearch中，您可以指定在索引期间使用一个分析器分析一个字段，并且当通过某个输入搜索时，该输入将由另一个分析器分析。所以你可以有这样的东西（从这里https://www.elastic.co/guide/en/elasticsearch/reference/current/search-analyzer.html）

{
  "mappings": {
    "properties": {
      "text": {
        "type": "text",
        "analyzer": "autocomplete",
        "search_analyzer": "standard"
      }
    }
  }
}

所以你的设置/映射变成了这样

{
  "settings": {
    "index": {
      "number_of_shards": "128",
      "number_of_replicas": "0",
      "analysis": {
        "filter": {
          "dbl_metaphone": {
            "type": "phonetic",
            "encoder": "double_metaphone",
            "max_code_len": 5
          },
          "given_synonyms": {
            "type": "synonym",
            "synonyms_path": "analysis/given_synonyms.txt"
          },
          "surname_synonyms": {
            "type": "synonym",
            "synonyms_path": "analysis/surname_synonyms.txt"
          }
        },
        "analyzer": {
          "dbl_metaphone": {
            "tokenizer": "standard",
            "filter": "dbl_metaphone"
          },
          "dbl_metaphone_given_synonym": {
            "tokenizer": "standard",
            "filter": [
              "given_synonyms",
              "dbl_metaphone"
            ]
          },
          "dbl_metaphone_surname_synonym": {
            "tokenizer": "standard",
            "filter": [
              "surname_synonyms",
              "dbl_metaphone"
            ]
          }
        }
      }
    }
  },
  "mappings": {
    "test": {
      "_all": {
        "enabled": false
      },
      "_source": {
        "enabled": true
      },
      "properties": {
        "GivenNames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_given_synonym"
            }
          }
        },
        "Surnames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_surname_synonym"
            }
          }
        },
        "FatherGivenNames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_given_synonym"
            }
          }
        },
        "FatherSurnames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_surname_synonym"
            }
          }
        },
        "MotherGivenNames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_given_synonym"
            }
          }
        },
        "MotherSurnames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_given_synonym"
            }
          }
        },
        "SpouseGivenNames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_given_synonym"
            }
          }
        },
        "SpouseSurnames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_surname_synonym"
            }
          }
        },
        "ChildrenGivenNames": {
          "type": "keyword",
          "index_options": "freqs",
          "store": "false",
          "similarity": "boolean",
          "norms": "false",
          "fields": {
            "phonetic": {
              "type": "text",
              "analyzer": "dbl_metaphone",
              "search_analyzer": "dbl_metaphone_given_synonym"
            }
          }
        },
        "BirthYears": {
          "type": "short"
        },
        "BirthLocations": {
          "type": "integer"
        },
        "DeathYears": {
          "type": "short"
        },
        "DeathLocations": {
          "type": "integer"
        },
        "MarriageLocations": {
          "type": "integer"
        },
        "MarriageYears": {
          "type": "integer"
        },
        "ResidenceLocations": {
          "type": "integer"
        }
      }
    }
  }
}

【讨论】：