Elasticsearch四种常见的相关度分数优化方法
**1、boost方式 **
简单粗暴，最常用。
需求：查询出title和content中包含java spark的document
方式1：
GET /forum/article/_search
{
"query": {
"bool": {
"should": [
{
"match": {
"title": "java spark"
}
},
{
"match": {
"content": "java spark"
}
}
]
}
}
}
结果：
{
"took": 3,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 0.970927,
"hits": [
{
"_index": "forum",
"_type": "article",
"_id": "5",
"_score": 0.970927,
"_source": {
"articleID": "DHJK-B-1395-#Ky5",
"userID": 3,
"hidden": false,
"postDate": "2017-03-01",
"tag": [
"elasticsearch"
],
"tag_cnt": 1,
"view_cnt": 10,
"title": "this is spark blog",
"content": "spark is best big data solution based on scala ,an programming language similar to java spark",
"sub_title": "haha, hello world",
"author_first_name": "Tonny",
"author_last_name": "Peter Smith",
"new_author_last_name": "Peter Smith",
"new_author_first_name": "Tonny"
}
},
{
"_index": "forum",
"_type": "article",
"_id": "2",
"_score": 0.8849759,
"_source": {
"articleID": "KDKE-B-9947-#kL5",
"userID": 1,
"hidden": false,
"postDate": "2017-01-02",
"tag": [
"java"
],
"tag_cnt": 1,
"view_cnt": 50,
"title": "this is java blog",
"content": "i think java is the best programming language",
"sub_title": "learned a lot of course",
"author_first_name": "Smith",
"author_last_name": "Williams",
"new_author_last_name": "Williams",
"new_author_first_name": "Smith"
}
},
]
}
}
方式2：
GET /forum/article/_search
{
"query": {
"bool": {
"should": [
{
"match": {
"title": {
"query": "java spark",
"boost" : 2
}
}
},
{
"match": {
"content": "java spark"
}
}
]
}
}
}
结果：
{
"took": 3,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total":2,
"max_score": 1.258609,
"hits": [
{
"_index": "forum",
"_type": "article",
"_id": "5",
"_score": 1.258609,
"_source": {
"articleID": "DHJK-B-1395-#Ky5",
"userID": 3,
"hidden": false,
"postDate": "2017-03-01",
"tag": [
"elasticsearch"
],
"tag_cnt": 1,
"view_cnt": 10,
"title": "this is spark blog",
"content": "spark is best big data solution based on scala ,an programming language similar to java spark",
"sub_title": "haha, hello world",
"author_first_name": "Tonny",
"author_last_name": "Peter Smith",
"new_author_last_name": "Peter Smith",
"new_author_first_name": "Tonny"
}
},
{
"_index": "forum",
"_type": "article",
"_id": "2",
"_score": 1.083544,
"_source": {
"articleID": "KDKE-B-9947-#kL5",
"userID": 1,
"hidden": false,
"postDate": "2017-01-02",
"tag": [
"java"
],
"tag_cnt": 1,
"view_cnt": 50,
"title": "this is java blog",
"content": "i think java is the best programming language",
"sub_title": "learned a lot of course",
"author_first_name": "Smith",
"author_last_name": "Williams",
"new_author_last_name": "Williams",
"new_author_first_name": "Smith"
}
}
]
}
}
两个结果对比很明显发现方式二比方式一的分数高出了好多。
2、重构查询结果
重构查询结果在ES新版本中，影响越来越小。一般情况没什么必要的话就别用了。
需求：查询出content中包含java spark solution beginner的document
方式1：
GET /forum/article/_search
{
"query": {
"bool": {
"should": [
{
"match": {
"content": "java"
}
},
{
"match": {
"content": "spark"
}
},
{
"match": {
"content": "solution"
}
},
{
"match": {
"content": "beginner"
}
}
]
}
}
}
结果：
{
"took": 3,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 1.113083,
"hits": [
{
"_index": "forum",
"_type": "article",
"_id": "4",
"_score": 1.113083,
"_source": {
"articleID": "QQPX-R-3956-#aD8",
"userID": 2,
"hidden": true,
"postDate": "2017-01-02",
"tag": [
"java",
"elasticsearch"
],
"tag_cnt": 2,
"view_cnt": 80,
"title": "this is java, elasticsearch, hadoop blog",
"content": "elasticsearch and hadoop are all very good solution, i am a beginner",
"sub_title": "both of them are good",
"author_first_name": "Robbin",
"author_last_name": "Li",
"new_author_last_name": "Li",
"new_author_first_name": "Robbin"
}
},
{
"_index": "forum",
"_type": "article",
"_id": "5",
"_score": 0.970927,
"_source": {
"articleID": "DHJK-B-1395-#Ky5",
"userID": 3,
"hidden": false,
"postDate": "2017-03-01",
"tag": [
"elasticsearch"
],
"tag_cnt": 1,
"view_cnt": 10,
"title": "this is spark blog",
"content": "spark is best big data solution based on scala ,an programming language similar to java spark",
"sub_title": "haha, hello world",
"author_first_name": "Tonny",
"author_last_name": "Peter Smith",
"new_author_last_name": "Peter Smith",
"new_author_first_name": "Tonny"
}
}
]
}
}
方式2：
GET /forum/article/_search
{
"query": {
"bool": {
"should": [
{
"match": {
"content": "java"
}
},
{
"match": {
"content": "spark"
}
},
{
"bool": {
"should": [
{
"match": {
"content": "solution"
}
},
{
"match": {
"content": "beginner"
}
}
]
}
}
]
}
}
}
结果：
{
"took": 3,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 1.113083,
"hits": [
{
"_index": "forum",
"_type": "article",
"_id": "4",
"_score": 1.113083,
"_source": {
"articleID": "QQPX-R-3956-#aD8",
"userID": 2,
"hidden": true,
"postDate": "2017-01-02",
"tag": [
"java",
"elasticsearch"
],
"tag_cnt": 2,
"view_cnt": 80,
"title": "this is java, elasticsearch, hadoop blog",
"content": "elasticsearch and hadoop are all very good solution, i am a beginner",
"sub_title": "both of them are good",
"author_first_name": "Robbin",
"author_last_name": "Li",
"new_author_last_name": "Li",
"new_author_first_name": "Robbin"
}
},
{
"_index": "forum",
"_type": "article",
"_id": "5",
"_score": 0.970927,
"_source": {
"articleID": "DHJK-B-1395-#Ky5",
"userID": 3,
"hidden": false,
"postDate": "2017-03-01",
"tag": [
"elasticsearch"
],
"tag_cnt": 1,
"view_cnt": 10,
"title": "this is spark blog",
"content": "spark is best big data solution based on scala ,an programming language similar to java spark",
"sub_title": "haha, hello world",
"author_first_name": "Tonny",
"author_last_name": "Peter Smith",
"new_author_last_name": "Peter Smith",
"new_author_first_name": "Tonny"
}
}
]
}
}
对比方式1和方式2的结果发现，相关度分数一模一样，几乎无差别。
3、negative boost
需求：搜索包含java，不包含spark的doc，但是这样很死板
GET /forum/article/_search
{
"query": {
"bool": {
"must": [
{
"match": {
"content": "java"
}
}
],
"must_not": [
{
"match": {
"content": "spark"
}
}
]
}
}
}
结果
{
"took": 58,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 1,
"max_score": 0.68640786,
"hits": [
{
"_index": "forum",
"_type": "article",
"_id": "2",
"_score": 0.68640786,
"_source": {
"articleID": "KDKE-B-9947-#kL5",
"userID": 1,
"hidden": false,
"postDate": "2017-01-02",
"tag": [
"java"
],
"tag_cnt": 1,
"view_cnt": 50,
"title": "this is java blog",
"content": "i think java is the best programming language",
"sub_title": "learned a lot of course",
"author_first_name": "Smith",
"author_last_name": "Williams",
"new_author_last_name": "Williams",
"new_author_first_name": "Smith"
}
}
]
}
}
只包含了java并且不包含spark的返回了。
需求：搜索包含java，尽量不包含spark的doc，如果包含了spark，不会说排除掉这个doc，而是说将这个doc的分数降低。
包含了nagative term的doc，分数乘以negative boost分数降低。
GET /forum/article/_search
{
"query": {
"boosting": {
"positive": {
"match": {
"content": "java"
}
},
"negative": {
"match": {
"content": "spark"
}
},
"negative_boost": 0.2
}
}
}
结果
{
"took": 10,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 2,
"max_score": 0.68640786,
"hits": [
{
"_index": "forum",
"_type": "article",
"_id": "2",
"_score": 0.68640786,
"_source": {
"articleID": "KDKE-B-9947-#kL5",
"userID": 1,
"hidden": false,
"postDate": "2017-01-02",
"tag": [
"java"
],
"tag_cnt": 1,
"view_cnt": 50,
"title": "this is java blog",
"content": "i think java is the best programming language",
"sub_title": "learned a lot of course",
"author_first_name": "Smith",
"author_last_name": "Williams",
"new_author_last_name": "Williams",
"new_author_first_name": "Smith"
}
},
{
"_index": "forum",
"_type": "article",
"_id": "5",
"_score": 0.05753642,
"_source": {
"articleID": "DHJK-B-1395-#Ky5",
"userID": 3,
"hidden": false,
"postDate": "2017-03-01",
"tag": [
"elasticsearch"
],
"tag_cnt": 1,
"view_cnt": 10,
"title": "this is spark blog",
"content": "spark is best big data solution based on scala ,an programming language similar to java spark",
"sub_title": "haha, hello world",
"author_first_name": "Tonny",
"author_last_name": "Peter Smith",
"new_author_last_name": "Peter Smith",
"new_author_first_name": "Tonny"
}
}
]
}
}
返回了两条，并且包含spark的那条数据score是0.0X，是因为会将得出的结果乘以0.2，降低了他的相关度分数，但是也会被检索出来
4、constant_score
如果你压根儿不需要相关度评分，直接走constant_score，所有doc分数都是1，没有评分的概念。
GET /forum/article/_search
{
"query": {
"bool": {
"should": [
{
"constant_score": {
"query": {
"match": {
"title": "java"
}
}
}
},
{
"constant_score": {
"query": {
"match": {
"title": "spark"
}
}
}
}
]
}
}
}
