Es 默认相关性算法 #
Es 5.0 之前，默认的相关性算分采用的是 TF-IDF，而之后则默认采用 BM25
用 Es 的 explain 查询可以看到详细的算分过程：
GET /your_index/_search
{
"query": {
"match_phrase": {"content": "婚姻纠纷"}
},
"fields": ["title"],
"_source": false,
"explain": true
}
输出：
{
...
"_explanation": {
"value": 25.720493,
"description": "weight(content:婚姻纠纷 in 2534301) [PerFieldSimilarity], result of:",
"details": [
{
"value": 25.720493,
"description": "score(freq=2.0), computed as boost * idf * tf from:",
"details": [
{
"value": 2.2,
"description": "boost",
"details": []
},
{
"value": 13.949452,
"description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
"details": [
{
"value": 1,
"description": "n, number of documents containing term",
"details": []
},
{
"value": 1714988,
"description": "N, total number of documents with field",
"details": []
}
]
},
{
"value": 0.838107,
"description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
"details": [
{
"value": 2,
"description": "freq, occurrences of term within document",
"details": []
},
{
"value": 1.2,
"description": "k1, term saturation parameter",
"details": []
},
{
"value": 0.75,
"description": "b, length normalization parameter",
"details": []
},
{
"value": 80,
"description": "dl, length of field (approximate)",
"details": []
},
{
"value": 834.0069,
"description": "avgdl, average length of field",
"details": []
}
]
}
]
}
]
}
},
文档分数 = boost * idf * tf idf = log(1 + (N - n + 0.5) / (n + 0.5)) tf = freq / (freq + k1 * (1 - b + b * dl / avgdl))
| 变量 | 解释 |
|---|---|
| boost | 权重分数，这个可以在查询语句中传入 |
| idf | 逆文档率 = log(1 + (N - n + 0.5) / (n + 0.5))N - 总的文档数n - 出现“联合”一次的文档数。 |
| tf | 词频 = freq / (freq + k1 * (1 - b + b * dl / avgdl)) freq - 搜索词在该文档出现的次数k1 - 控制词频影响的重要程度，默认值 1.2b - 控制对大长文本字段的惩罚程度，默认值 0.75dl - 该文档中检索字段的长度avgdl - 检索字段的平均长度。 |
BM25 调参 #
调整方式，只能在创建索引的时候调整，无法动态调整。
创建自定义的 custom_similarity，参数 b 的值设置为 0，k1 保持默认的 1.2，当 b = 0 时，tf 公式就变成了 freq / (freq + k1)，tf 分数完全忽略文本长度的影响。
PUT /your_index
{
"settings": {
"similarity": {
"custom_similarity": {
"type": "BM25",
"b": "0",
"k1": "1.2"
}
}
},
"mappings": {
"properties": {
"id": {
"type": "integer",
"index": true
},
"title": {
"type": "text",
"index": true,
"similarity": "custom_similarity"
}
}
}
}
但是 idf 中 的 N 和 n 取决于搜索词和数据，无法修改，只能调 tf 的整参数 b 和 k1，无法完全消除文本长度对相关性的影响。
自定义 similarity 算法 #
建立索引，自定义 similarity：
PUT /index
{
"settings": {
"number_of_shards": 1,
"similarity": {
"scripted_tfidf": {
"type": "scripted",
"script": {
"source": "double tf = Math.sqrt(doc.freq); double idf = Math.log((field.docCount+1.0)/(term.docFreq+1.0)) + 1.0; double norm = 1/Math.sqrt(doc.length); return query.boost * tf * idf * norm;"
}
}
}
},
"mappings": {
"properties": {
"field": {
"type": "text",
"similarity": "scripted_tfidf"
}
}
}
}
导入一些数据：
PUT /index/_doc/1
{
"field": "foo bar foo"
}
PUT /index/_doc/2
{
"field": "bar baz"
}
POST /index/_refresh
搜索测试：
GET /index/_search?explain=true
{
"query": {
"query_string": {
"query": "foo^1.7",
"default_field": "field"
}
}
}
搜索结果：
{
"took": 12,
"timed_out": false,
"_shards": {
"total": 1,
"successful": 1,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 1,
"relation": "eq"
},
"max_score": 1.9508477,
"hits": [
{
"_shard": "[index][0]",
"_node": "OzrdjxNtQGaqs4DmioFw9A",
"_index": "index",
"_id": "1",
"_score": 1.9508477,
"_source": {
"field": "foo bar foo"
},
"_explanation": {
"value": 1.9508477,
"description": "weight(field:foo in 0) [PerFieldSimilarity], result of:",
"details": [
{
"value": 1.9508477,
"description": "score from ScriptedSimilarity(weightScript=[null], script=[Script{type=inline, lang='painless', idOrCode='double tf = Math.sqrt(doc.freq); double idf = Math.log((field.docCount+1.0)/(term.docFreq+1.0)) + 1.0; double norm = 1/Math.sqrt(doc.length); return query.boost * tf * idf * norm;', options={}, params={}}]) computed from:",
"details": [
{
"value": 1.0,
"description": "weight",
"details": []
},
{
"value": 1.7,
"description": "query.boost",
"details": []
},
{
"value": 2,
"description": "field.docCount",
"details": []
},
{
"value": 4,
"description": "field.sumDocFreq",
"details": []
},
{
"value": 5,
"description": "field.sumTotalTermFreq",
"details": []
},
{
"value": 1,
"description": "term.docFreq",
"details": []
},
{
"value": 2,
"description": "term.totalTermFreq",
"details": []
},
{
"value": 2.0,
"description": "doc.freq",
"details": []
},
{
"value": 3,
"description": "doc.length",
"details": []
}
]
}
]
}
}
]
}
}
可以使用的值已经在上面的 description 里面列举出来。