提问：布和纸怕什么？

关于ES的评分？

Elasticsearch | 作者 zev | 发布于2018年08月23日 | 阅读数：3495
分享到：QQ空间新浪微博微信QQ好友印象笔记有道云笔记
搜索关键词: 关于
 
第一个结果和第四个结果中‘单位’都有匹配，
但第四个结果‘标题’也有匹配，正常要得结果肯定是在前面的。
是什么原因导致它的评分降低？求解
是查询语句问题？还是设置什么参数没设置好？

2.png

 
 
GET document/_search?explain=true

{

  "query": {

  "dis_max" : {

    "tie_breaker" : 0.4,

    "queries" : [

      {

        "match" : {

          "文件标题" : {

            "query" : "关于",

            "operator" : "OR",

            "prefix_length" : 0,

            "max_expansions" : 50,

            "fuzzy_transpositions" : true,

            "lenient" : false,

            "zero_terms_query" : "NONE",

            "auto_generate_synonyms_phrase_query" : true,

            "boost" : 5.0

          }

        }

      },

      {

        "match" : {

          "单位" : {

            "query" : "关于",

            "operator" : "OR",

            "prefix_length" : 0,

            "max_expansions" : 50,

            "fuzzy_transpositions" : true,

            "lenient" : false,

            "zero_terms_query" : "NONE",

            "auto_generate_synonyms_phrase_query" : true,

            "boost" : 1.0

          }

        }

      },

      {

        "match" : {

          "类型" : {

            "query" : "关于",

            "operator" : "OR",

            "prefix_length" : 0,

            "max_expansions" : 50,

            "fuzzy_transpositions" : true,

            "lenient" : false,

            "zero_terms_query" : "NONE",

            "auto_generate_synonyms_phrase_query" : true,

            "boost" : 1.0

          }

        }

      },

      {

        "match" : {

          "意见" : {

            "query" : "关于",

            "operator" : "OR",

            "prefix_length" : 0,

            "max_expansions" : 50,

            "fuzzy_transpositions" : true,

            "lenient" : false,

            "zero_terms_query" : "NONE",

            "auto_generate_synonyms_phrase_query" : true,

            "boost" : 1.0

          }

        }

      },

      {

        "match" : {

          "来文号" : {

            "query" : "关于",

            "operator" : "OR",

            "prefix_length" : 0,

            "max_expansions" : 50,

            "fuzzy_transpositions" : true,

            "lenient" : false,

            "zero_terms_query" : "NONE",

            "auto_generate_synonyms_phrase_query" : true,

            "boost" : 1.0

          }

        }

      },

      {

        "match" : {

          "来文编号" : {

            "query" : "关于",

            "operator" : "OR",

            "prefix_length" : 0,

            "max_expansions" : 50,

            "fuzzy_transpositions" : true,

            "lenient" : false,

            "zero_terms_query" : "NONE",

            "auto_generate_synonyms_phrase_query" : true,

            "boost" : 1.0

          }

        }

      }

    ],

    "boost" : 1.0

  }

  }

}
 
"mappings": {



    "doc": {

        "properties": {

            "单位": {

                "analyzer": "ik_smart",

                "type": "text"

            },

            "公文类型": {

                "type": "text","fields": {

                    "keyword": {

                        "ignore_above": 256,"type": "keyword"}}},

            "来文号": {

                "analyzer": "ik_smart",

                "type": "text"

            },

            "办结日期": {

                "type": "long"

            },

            "登记日期": {

                "type": "long"

            },

            "文件标题": {

                "analyzer": "ik_smart",

                "type": "text"

            },

            "意见": {

                "analyzer": "ik_smart",

                "type": "text"

            },

            "来文编号": {

                "analyzer": "ik_smart",

                "type": "text"

            },

            "类型": {

                "analyzer": "ik_smart",

                "type": "text"

            }

        }

    }



},

 
2018-08-23 添加评论

没有找到相关结果

    已邀请:

    medcl - 今晚打老虎。

    赞同来自: derobukal

    打分结果已经告诉你了，出问题的不是条件，是你的关键字，“关于”这个词在标题里面是高频词，差不多一半的标题里面都出现了“关于”，所以，即使命中了，贡献的分值也不高，因为“关于”这个词实在是普通。而在“单位”字段里面就比较少见了，所以，单位字段命中之后贡献的分值很大。
    2018-09-03 1 0
    分享

    zev

    赞同来自:

    ​这是两个结果的评分：
          {
    
        "_shard": "[document_v2][0]",
    
        "_node": "DmPg1E3mR3KTfzUpqVzPDQ",
    
        "_index": "document_v2",
    
        "_type": "doc",
    
        "_id": "g0102_bc491024-fb6b-469d-bb3c-68b144fc988f",
    
        "_score": 6.2364345,
    
        "_source": {
    
          "文件标题": "区党委",
    
          "单位": "自治区党委转发中央组织部关于姜兴和同志退休的通知",
    
        },
    
        "_explanation": {
    
          "value": 6.2364345,
    
          "description": "max plus 0.4 times others of:",
    
          "details": [
    
            {
    
              "value": 6.2364345,
    
              "description": "weight(单位:关于 in 12443) [PerFieldSimilarity], result of:",
    
              "details": [
    
                {
    
                  "value": 6.2364345,
    
                  "description": "score(doc=12443,freq=1.0 = termFreq=1.0\n), product of:",
    
                  "details": [
    
                    {
    
                      "value": 9.842326,
    
                      "description": "idf, computed as log(1 + (docCount - docFreq + 0.5) / (docFreq + 0.5)) from:",
    
                      "details": [
    
                        {
    
                          "value": 3,
    
                          "description": "docFreq",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 65846,
    
                          "description": "docCount",
    
                          "details": []
    
                        }
    
                      ]
    
                    },
    
                    {
    
                      "value": 0.6336342,
    
                      "description": "tfNorm, computed as (freq * (k1 + 1)) / (freq + k1 * (1 - b + b * fieldLength / avgFieldLength)) from:",
    
                      "details": [
    
                        {
    
                          "value": 1,
    
                          "description": "termFreq=1.0",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 1.2,
    
                          "description": "parameter k1",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 0.75,
    
                          "description": "parameter b",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 4.557938,
    
                          "description": "avgFieldLength",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 11,
    
                          "description": "fieldLength",
    
                          "details": []
    
                        }
    
                      ]
    
                    }
    
                  ]
    
                }
    
              ]
    
            }
    
          ]
    
        }
    
      },
    
            {
    
        "_shard": "[document_v2][0]",
    
        "_node": "DmPg1E3mR3KTfzUpqVzPDQ",
    
        "_index": "document_v2",
    
        "_type": "doc",
    
        "_id": "g0102_038731a7-1ce2-430c-a522-4baba857ac2c",
    
        "_score": 5.0754642,
    
        "_source": {
    
          "文件标题": "关于在南宁建设周氏兄弟国际艺术谷的设想",
    
          "单位": "周红波市长在《关于南宁建设周氏兄弟国际艺术谷的设想》的批示件",
    
          "类型": "收文",
    
        },
    
        "_explanation": {
    
          "value": 5.0754642,
    
          "description": "max plus 0.4 times others of:",
    
          "details": [
    
            {
    
              "value": 1.0641516,
    
              "description": "weight(文件标题:关于 in 16401) [PerFieldSimilarity], result of:",
    
              "details": [
    
                {
    
                  "value": 1.0641516,
    
                  "description": "score(doc=16401,freq=1.0 = termFreq=1.0\n), product of:",
    
                  "details": [
    
                    {
    
                      "value": 5,
    
                      "description": "boost",
    
                      "details": []
    
                    },
    
                    {
    
                      "value": 0.19654904,
    
                      "description": "idf, computed as log(1 + (docCount - docFreq + 0.5) / (docFreq + 0.5)) from:",
    
                      "details": [
    
                        {
    
                          "value": 87490,
    
                          "description": "docFreq",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 106492,
    
                          "description": "docCount",
    
                          "details": []
    
                        }
    
                      ]
    
                    },
    
                    {
    
                      "value": 1.0828357,
    
                      "description": "tfNorm, computed as (freq * (k1 + 1)) / (freq + k1 * (1 - b + b * fieldLength / avgFieldLength)) from:",
    
                      "details": [
    
                        {
    
                          "value": 1,
    
                          "description": "termFreq=1.0",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 1.2,
    
                          "description": "parameter k1",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 0.75,
    
                          "description": "parameter b",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 14.760095,
    
                          "description": "avgFieldLength",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 12,
    
                          "description": "fieldLength",
    
                          "details": []
    
                        }
    
                      ]
    
                    }
    
                  ]
    
                }
    
              ]
    
            },
    
            {
    
              "value": 4.6498036,
    
              "description": "weight(单位:关于 in 16401) [PerFieldSimilarity], result of:",
    
              "details": [
    
                {
    
                  "value": 4.6498036,
    
                  "description": "score(doc=16401,freq=1.0 = termFreq=1.0\n), product of:",
    
                  "details": [
    
                    {
    
                      "value": 9.842326,
    
                      "description": "idf, computed as log(1 + (docCount - docFreq + 0.5) / (docFreq + 0.5)) from:",
    
                      "details": [
    
                        {
    
                          "value": 3,
    
                          "description": "docFreq",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 65846,
    
                          "description": "docCount",
    
                          "details": []
    
                        }
    
                      ]
    
                    },
    
                    {
    
                      "value": 0.47242934,
    
                      "description": "tfNorm, computed as (freq * (k1 + 1)) / (freq + k1 * (1 - b + b * fieldLength / avgFieldLength)) from:",
    
                      "details": [
    
                        {
    
                          "value": 1,
    
                          "description": "termFreq=1.0",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 1.2,
    
                          "description": "parameter k1",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 0.75,
    
                          "description": "parameter b",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 4.557938,
    
                          "description": "avgFieldLength",
    
                          "details": []
    
                        },
    
                        {
    
                          "value": 17,
    
                          "description": "fieldLength",
    
                          "details": []
    
                        }
    
                      ]
    
                    }
    
                  ]
    
                }
    
              ]
    
            }
    
          ]
    
        }
    
      }
    2018-08-23 0 0
    分享

    yayg2008

    赞同来自:

    根据打分结果，仔细分析一下，就可以知道哪个打分项占的比重最大；再结合BM25的参数因子进行调整，验证。
    2018-08-28 0 0
    分享
    为什么被折叠? 0 个回复被折叠

    要回复问题请先登录或注册

    Copyright © 2024 · CC BY-NC-SA 3.0

    本站服务器及带宽由 提供赞助