elasticsearch创建mapping字段索引说明

╰+攻爆jí腚メ 2022-02-20 03:41 466阅读 0赞

本样例版本为6.x  
7.x中去掉了type，但是type依然存在，为默认值：\_doc

## 样例 ##

PUT paper
    {
    	"settings": {
    		"number_of_replicas": 0,
    		"number_of_shards": 2
    	},
    	"mappings": {
    		"pap": {
    			"properties": {
    				"linkCount": {
    					"fielddata": true,
    					"store": true,
    					"type": "text"
    				},
    				"pubDate": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text"
    				},
    				"publish_date": {
    					"format": "yyyy-MM-dd HH:mm:ss||yyyy-MM-dd||epoch_millis",
    					"ignore_malformed": true,
    					"type": "date"
    				},
    				"source": {
    					"copy_to": "fullcontent",
    					"fielddata": true,
    					"store": true,
    					"type": "text"
    				},
    				"summary": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text",
    					"similarity": "BM25",
    					"index_options": "offsets",
    					"analyzer": "ik_max_word"
    				},
    				"title": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text",
    					"similarity": "BM25",
    					"index_options": "offsets",
    					"analyzer": "ik_max_word"
    				},
    				"url": {
    					"type": "text"
    				},
    				"viewCount": {
    					"fielddata": true,
    					"store": true,
    					"type": "text"
    				},
    				"year": {
    					"fielddata": true,
    					"store": true,
    					"type": "text"
    				},
    				"columnName": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text",
    					"similarity": "BM25",
    					"index_options": "offsets"
    				},
    				"doi": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text"
    				},
    				"downloadCount": {
    					"fielddata": true,
    					"store": true,
    					"type": "text"
    				},
    				"enTitle": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text",
    					"similarity": "BM25",
    					"analyzer": "ik_max_word"
    				},
    				"id": {
    					"store": true,
    					"type": "keyword"
    				},
    				"journal": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text",
    					"similarity": "BM25",
    					"index_options": "offsets",
    					"analyzer": "ik_max_word"
    				},
    				"keyWords": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "keyword"
    				},
    				"authors": {
    					"properties": {
    						"author": {
    							"copy_to": "fullcontent",
    							"type": "keyword"
    						},
    						"doctorId": {
    							"store": true,
    							"type": "keyword"
    						},
    						"hospitalName": {
    							"copy_to": "fullcontent",
    							"store": true,
    							"type": "text",
    							"similarity": "BM25",
    							"index_options": "offsets",
    							"analyzer": "ik_max_word"
    						},
    						"doctor_hcoid": {
    							"store": true,
    							"type": "keyword"
    						},
    						"doctor_hcpid": {
    							"store": true,
    							"type": "keyword"
    						},
    						"institution": {
    							"copy_to": "fullcontent",
    							"store": true,
    							"type": "text",
    							"similarity": "BM25",
    							"index_options": "offsets",
    							"analyzer": "ik_max_word"
    						},
    						"url": {
    							"type": "text"
    						}
    					}
    				},
    				"fullcontent": {
    					"copy_to": "fullcontent",
    					"store": true,
    					"type": "text",
    					"similarity": "BM25",
    					"index_options": "offsets",
    					"analyzer": "ik_max_word"
    				}
    			}
    		}
    	}
    }

## setting： ##

"number_of_replicas": 0,		// 备份数
    "number_of_shards": 2		//分片数

## type： ##

字段类型概述
    一级分类	二级分类	具体类型
    核心类型	字符串类型	text,keyword
    整数类型	integer,long,short,byte
    浮点类型	double,float,half_float,scaled_float
    逻辑类型	boolean
    日期类型	date
    范围类型	range
    二进制类型	binary
    复合类型	数组类型	array
    对象类型	object
    嵌套类型	nested
    地理类型	地理坐标类型	geo_point
    地理地图	geo_shape
    特殊类型	IP类型	ip
    范围类型	completion
    令牌计数类型	token_count
    附件类型	attachment
    抽取类型	percolator

## type:text ##

例：

"summary": {
    	"type": "text",
    	"analyzer": "ik_max_word"
    }

此mapping表示：summary字段为字符串，分词器采用**全词匹配**。若希望采用分词匹配，则mapping应为：

"summary": {
    	"type": "text",
    	"analyzer": "ik_smart"
    }

## type:keyword ##

例：

"cnName": {
    	"store": true,
    	"type": "keyword"
    },

此mapping表示：cnName字段为字符串，类型为**keyword**，为**全词匹配**，适合**精确匹配查找**，支持**groupby**。

## 字段需要groupby：“fielddata”: true ##

"linkCount": {
    	"fielddata": true,
    	"store": true,
    	"type": "text"
    }

这个字段需要 **groupby** 且 **type** 为 **text** 的时候，**必须** 将 **fielddata** 设置为 **true**

## “store”: true ##

elasticsearch将字段保存一份源文档到 \_source

## 时间格式 ##

"publish_date": {
    	"format": "yyyy-MM-dd HH:mm:ss||yyyy-MM-dd||epoch_millis",
    	"ignore_malformed": true,
    	"type": "date"
    }

ignore\_malformed：取值为true或false，默认值是false。若要忽略**格式错误**的数值，则应设置为true。

## elasticsearch 中的相似度模型：“similarity”: “BM25”, ##

例：

"summary": {
    	"copy_to": "fullcontent",
    	"store": true,
    	"type": "text",
    	"similarity": "BM25",
    	"index_options": "offsets",
    	"analyzer": "ik_max_word"
    },

此mapping中的 **“similarity”: “BM25”** 为了避免**搜索词**在该字段中出现的频率过高而影响评分。  
比如：我们搜索fire fox，假如返回两篇文章 doc1 和 doc2，doc1 的评分为15，doc2的评分为10。但是，有可能doc1是一篇很长的关于火灾的文章；而doc2则是一篇关于firefox浏览器的使用教程。而我们的预期显然则是更偏向于后者，此时则需要在mapping中加入**相似度模型**。  
关于BM25的理论基础：推荐阅读https://www.elastic.co/guide/cn/elasticsearch/guide/current/pluggable-similarites.html