进阶-第27__深度探秘搜索技术_实战用function_score自定义相关度分数算法

我们可以做到自定义一个function_score函数，自己将某个field的值，跟es内置算出来的分数进行运算，然后由自己指定的field来进行分数的增强

添加字段和数据

给所有的帖子数据增加follower数量

POST /forum/article/_bulk { "update": { "_id": "1"} } { "doc" : {"follower_num" : 5} } { "update": { "_id": "2"} } { "doc" : {"follower_num" : 10} } { "update": { "_id": "3"} } { "doc" : {"follower_num" : 25} } { "update": { "_id": "4"} } { "doc" : {"follower_num" : 3} } { "update": { "_id": "5"} } { "doc" : {"follower_num" : 60} }

 

将对帖子搜索得到的分数，跟follower_num进行运算，由follower_num在一定程度上增强帖子的分数

看帖子的人越多，那么帖子的分数就越高

测试-不和点击量挂钩

GET /forum/article/_search {       "query": {         "multi_match": {           "query": "java spark",           "fields": ["tile", "content"]         }     } } 结果： {   "took": 2,   "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",           "follower_num": 10         }       },       {         "_index": "forum",         "_type": "article",         "_id": "5",         "_score": 0.68324494,         "_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",           "follower_num": 60         }       }     ]   } }

测试-和点击量挂钩

GET /forum/article/_search {   "query": {     "function_score": {       "query": {         "multi_match": {           "query": "java spark",           "fields": ["tile", "content"]         }       },       "field_value_factor": {         "field": "follower_num",//影响分数的字段         "modifier": "log1p",//对影响分数的字段进行log(1 + number_of_votes)         "factor": 0.5// * log(1 + factor * number_of_votes) 对应性分数的字段进行乘以影响因子       },       "boost_mode": "multiply",//员分数和影响分数字段的关系old_score * log(1 + factor * number_of_votes),如果不写：默认的是相乘       "max_boost": 2//最大分数的限制     }   } } 结果： {   "took": 2,   "timed_out": false,   "_shards": {     "total": 5,     "successful": 5,     "failed": 0   },   "hits": {     "total": 2,     "max_score": 1.0189654,     "hits": [       {         "_index": "forum",         "_type": "article",         "_id": "5",         "_score": 1.0189654,         "_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",           "follower_num": 60         }       },       {         "_index": "forum",         "_type": "article",         "_id": "2",         "_score": 0.53412914,         "_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",           "follower_num": 10         }       }     ]   } }

 

 

如果只有field(比如:follower_num)，那么会将每个doc的分数都乘以follower_num，如果有的doc follower是0，那么分数就会变为0，效果很不好。因此一般会加个log1p函数，公式会变为，new_score = old_score * log(1 + number_of_votes)，这样出来的分数会比较合理

再加个factor，可以进一步影响分数，new_score = old_score * log(1 + factor * number_of_votes)

boost_mode(如果不写：默认的是相乘)，可以决定分数与指定字段的值如何计算，multiply，sum，min，max，replace

max_boost，限制计算出来的分数不要超过max_boost指定的值(这个参数越来越没结果了)