理解Elasticsearch的相关性

Demo

搜索名字为Sophia zhou的文档

// request body
{
    "query":{
        "match":{
            "name":"Sophia zhou"
        }
    }
}

响应内容,可以很明显看到Sophia Wu和John Zhou都被查询出来了，对应的score如下

name	_score
Sophia Zhou	0.5753642
Sophia Wu	0.2876821
John Zhou	0.2876821

{
    "took": 2,
    "timed_out": false,
    "_shards": {
        "total": 5,
        "successful": 5,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        "total": 3,
        "max_score": 0.5753642,
        "hits": [
            {
                "_index": "corp",
                "_type": "employee",
                "_id": "2",
                "_score": 0.5753642,
                "_source": {
                    "country": "CA",
                    "name": "Sophia Zhou",
                    "hobbies": [
                        "foods",
                        "read"
                    ],
                    "age": 15
                }
            },
            {
                "_index": "corp",
                "_type": "employee",
                "_id": "1",
                "_score": 0.2876821,
                "_source": {
                    "country": "US",
                    "name": "Sophia Wu",
                    "birthday": "1631096936000",
                    "hobbies": [
                        "foods",
                        "play games",
                        "read",
                        "talk"
                    ],
                    "age": 1
                }
            },
            {
                "_index": "corp",
                "_type": "employee",
                "_id": "3",
                "_score": 0.2876821,
                "_source": {
                    "country": "CA",
                    "name": "John Zhou",
                    "hobbies": [
                        "swim",
                        "read"
                    ],
                    "birthday": 1315477736000,
                    "age": 11
                }
            }
        ]
    }
}

解释

可以理解为ES作为全文搜索引擎，会使用分词策略进行查询，Sophia zhou 会被分词为Sophia、zhou两个词，同理三条文档中的name都会被分词，例如John Zhou会被分词为John、Zhou；经过分词后ES会计算文档与查询条件的相关性，查询结果默认根据相关性倒序排列。这也就是ES适合做搜索引擎的原因之一。