Opensearch中支持模糊匹配的cross_fields替代方案

我需要实现一个面向学生的职位搜索功能，要求基于职位关键词查询工作。目前用cross_fields类型的multi_match查询实现了理想的排序结果（多字段匹配权重累加，跨字段满足所有查询词匹配），但cross_fields不支持模糊匹配，无法处理拼写错误的场景。

测试数据集 #

POST _bulk
{"index":{"_index":"duarte-search-role","_id":"Job 1"}}
{"title":"Marine Biologist 1","overview":"Marine Biologist","opportunity_type_name":"Graduate Job","expired":false}
{"index":{"_index":"duarte-search-role","_id":"Job 2"}}
{"title":"Marine Biologist 2","overview":"No keyword","opportunity_type_name":"Graduate Job","expired":false}
{"index":{"_index":"duarte-search-role","_id":"Job 3"}}
{"title":"Comparison Job 3","overview":"Marine Biologist","opportunity_type_name":"Graduate Job","expired":false}
{"index":{"_index":"duarte-search-role","_id":"Job 4"}}
{"title":"Marine Biologist 4","overview":"Marine Biologist","opportunity_type_name":"Internship","expired":false}
{"index":{"_index":"duarte-search-role","_id":"Job 5"}}
{"title":"Marine Biologist 5","overview":"No keyword","opportunity_type_name":"Internship","expired":false}
{"index":{"_index":"duarte-search-role","_id":"Job 6"}}
{"title":"Comparison Job 6","overview":"Marine Biologist","opportunity_type_name":"Internship","expired":false}

预期搜索逻辑 #

• 输入Marine Biologist：Job 1（标题+概述均匹配）排第一，Job 3（概述匹配）次之，Job 2（仅标题匹配）随后，以此类推
• 输入Graduate Marine Biologist：Job 1（标题+概述匹配关键词，机会类型匹配Graduate）排第一，Job 2次之，以此类推
• 输入Marine Biologist Internship：Job 4（标题+概述匹配关键词，机会类型匹配Internship）排第一，Job 5次之，以此类推

当前实现的查询（无模糊匹配） #

GET /duarte-search-role/_search?search_type=dfs_query_then_fetch
{ 
  "query": {
    "bool": {
      "must": [
        {
          "multi_match": {
            "query": "Marine Biologist Internship",
            "fields": [
              "title^100",
              "overview^50",
              "opportunity_type_name^30"
            ],
            "operator": "and",
            "type": "cross_fields",
            "tie_breaker": 1
          }
        }
      ],
      "filter": [
        {
          "term": {
            "expired": false
          }
        }
      ]
    }
  },
  "sort": [
    {
      "_score": {
        "order": "desc"
      }
    },
    {
      "application_close_date": {
        "order": "asc"
      }
    }
  ],
  "from": 0,
  "size": 8
}

核心需求 #

保留原有排序逻辑（多字段权重累加、跨字段满足所有查询词匹配）的同时，支持模糊匹配以处理拼写错误。

由于cross_fields类型不支持模糊匹配，可通过以下两种方案替代，同时保留原有排序逻辑：

方案1：使用query_string查询（无需拆分查询词） #

query_string支持跨字段模糊匹配，且可通过default_operator: AND确保所有查询词都被匹配，得分会根据字段权重累加，符合原有排序逻辑。

GET /duarte-search-role/_search?search_type=dfs_query_then_fetch
{ 
  "query": {
    "bool": {
      "must": [
        {
          "query_string": {
            "query": "Marine~ Biologist~ Internship~",
            "fields": [
              "title^100",
              "overview^50",
              "opportunity_type_name^30"
            ],
            "default_operator": "AND",
            "fuzziness": "AUTO"
          }
        }
      ],
      "filter": [
        {
          "term": {
            "expired": false
          }
        }
      ]
    }
  },
  "sort": [
    {
      "_score": {
        "order": "desc"
      }
    },
    {
      "application_close_date": {
        "order": "asc"
      }
    }
  ],
  "from": 0,
  "size": 8
}

说明 #

• 每个词后的~标记开启模糊匹配，也可通过全局fuzziness: "AUTO"统一设置模糊度（自动根据词长调整编辑距离）
• default_operator: AND确保所有查询词都必须在某个字段匹配到，模拟cross_fields的operator: AND效果
• 字段权重设置与原有查询一致，得分会根据匹配字段的权重累加，保留原有排序逻辑

方案2：拆分查询词，使用bool+multi_match组合（更灵活） #

将用户输入的查询拆分为单个关键词，每个词对应一个multi_match模糊查询，放入bool的must子句中，确保所有词都被匹配，同时保留字段权重和累加得分逻辑。

GET /duarte-search-role/_search?search_type=dfs_query_then_fetch
{ 
  "query": {
    "bool": {
      "must": [
        {
          "multi_match": {
            "query": "Marine",
            "fields": ["title^100", "overview^50", "opportunity_type_name^30"],
            "fuzziness": "AUTO"
          }
        },
        {
          "multi_match": {
            "query": "Biologist",
            "fields": ["title^100", "overview^50", "opportunity_type_name^30"],
            "fuzziness": "AUTO"
          }
        },
        {
          "multi_match": {
            "query": "Internship",
            "fields": ["title^100", "overview^50", "opportunity_type_name^30"],
            "fuzziness": "AUTO"
          }
        }
      ],
      "filter": [
        {
          "term": {
            "expired": false
          }
        }
      ]
    }
  },
  "sort": [
    {
      "_score": {
        "order": "desc"
      }
    },
    {
      "application_close_date": {
        "order": "asc"
      }
    }
  ],
  "from": 0,
  "size": 8
}

说明 #

• 需要在应用层将用户输入的查询拆分为单个词（可通过空格分割，注意处理连续空格等情况）
• 每个multi_match针对单个词进行跨字段模糊匹配，must子句确保所有词都必须匹配
• 字段权重与原有查询一致，得分由多个匹配字段的权重累加，完全保留原有排序逻辑
• 相比query_string，此方案更灵活，可针对不同词设置不同的模糊度或字段范围

内容的提问来源于stack exchange，提问作者kyuubi