【elasticsearch】Term vector笔记

Term vecotr：获取document中某个field内各个term的统计信息

    index-time: mapping配置、建立索引时生成term和field信息

    query-time：查看term vector时 现场计算统计信息，返回

term information:    term frequency in the field	term positions:       start and end offsets
term statistics: 设置term_statistics=true;	total term frequency, tfc  一个term在所有document中出现的频率
document frequency，有多少document包含这个term	field statistics: document count，有多少document包含这个field
sum of document frequency，一个field中所有term的df之和;	sum of total term frequency，一个field中的所有term的tf之和
index-time PUT /my_index {   "mappings": {     "my_type": {       "properties": {         "text": {             "type": "text",             "term_vector": "with_positions_offsets_payloads",             "store" : true,             "analyzer" : "fulltext_analyzer"          },          "fullname": {             "type": "text",             "analyzer" : "fulltext_analyzer"         }       }     }   },   "settings" : {     "index" : {       "number_of_shards" : 1,       "number_of_replicas" : 0     },     "analysis": {       "analyzer": {         "fulltext_analyzer": {           "type": "custom",           "tokenizer": "whitespace",           "filter": [             "lowercase",             "type_as_payload"           ]         }       }     }   } }	GET /my_index/my_type/1/_termvectors {   "fields" : ["text"],   "offsets" : true,   "payloads" : true,   "positions" : true,   "term_statistics" : true,   "field_statistics" : true } 一个term出现一次就是一个token 出现的位置start_offset 手动指定doc的term vector 4、GET /my_index/my_type/_termvectors {   "doc" : {     "fullname" : "Leo Li",     "text" : "hello test test test"   },   "fields" : ["text"],   "offsets" : true,   "payloads" : true,   "positions" : true,   "term_statistics" : true,   "field_statistics" : true } 手动指定一个doc，如上的“text”， 将term分词，对每个term，计算它在先有的doc中的统计信息 4、中最后的}上添加分词器 "per_field_analyzer" : {     "text": "standard"   }
GET /my_index/my_type/_termvectors { ……   "filter" : {       "max_num_terms" : 3,最多terms个数       "min_term_freq" : 1,最少term频率       "min_doc_freq" : 1最少doc中出现次数     } } 4、中添加terms filter； 根据term统计信息，过滤出你想要看到的term vector统计结果 滤掉一些出现频率过低的term	multi term vector GET _mtermvectors {    "docs": [       {          "_index": "my_index",          "_type": "my_type",          "_id": "2",       "term_statistics": true       },       {          "_index": "my_index",          "_type": "my_type",          "_id": "1",          "fields": [             "text"          ]       }    ] } GET /my_index/_mtermvectors {    "docs": [       {          "_type": "test",          "_id": "2",          "fields": [             "text"          ],          "term_statistics": true       },       {          "_type": "test",          "_id": "1"       }    ] } GET /my_index/my_type/_mtermvectors {    "docs": [       {          "_id": "2",          "fields": [             "text"          ],          "term_statistics": true       },       {          "_id": "1"       }    ] } GET /_mtermvectors {    "docs": [       {          "_index": "my_index",          "_type": "my_type",          "doc" : {             "fullname" : "Leo Li",             "text" : "hello test test test"          }       },       {          "_index": "my_index",          "_type": "my_type",          "doc" : {            "fullname" : "Leo Li",            "text" : "other hello test ..."          }       }    ] }