ElasticSearch学习笔记-ngram、中文拼音、简繁体搜索记录_java整合elasticsearch 7.17.5 实现中文+拼音+简繁搜索-优快云博客

本文链接：https://blog.youkuaiyun.com/fighting_one_piece/article/details/100516819

本文记录了在ElasticSearch 7.3.0版本中进行中文搜索的相关实践，包括ngram分词器的使用、中文拼音处理以及支持简繁体搜索的配置。通过创建Mapping，插入数据并验证分词及查询效果，展示了ElasticSearch在中文检索方面的功能。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

ElasticSearch版本：elasticsearch-7.3.0

ElasticSearch相关插件安装可以参考：

ElasticSearch学习笔记-插件安装记录_人生偌只如初见的博客-优快云博客

ElasticSearch服务上创建Mapping

curl -H "Content-Type: application/json" -XPUT 'http://192.168.0.1:9200/book_v2/' -d '
{
  "settings": {
    "analysis": {
      "analyzer": {
        "pinyin_analyzer_1": {
          "tokenizer": "pinyin_tokenizer_1"
        },
        "pinyin_analyzer_2": {
          "tokenizer": "pinyin_tokenizer_2"
        },
        "tsconvert": {
          "tokenizer": "tsconvert_tokenizer",
          "filter": [
            "tsconvert_filter"
          ]
        },
        "autocomplete_analyzer": {
          "type": "custom",
          "tokenizer": "ik_smart",
          "filter": [
            "autocomplete_filter"
          ]
        },
        "custom_analyzer_1": {
          "type": "custom",
          "tokenizer": "ik_smart",
          "filter": [
            "cus_pinyin_filter_1"
          ]
        },
        "custom_analyzer_2": {
          "type": "custom",
          "tokenizer": "ik_max_word",
          "filter": [
            "stconvert_filter"
          ]
        },
        "custom_analyzer_3": {
          "type": "custom",
          "tokenizer": "ik_max_word",
          "filter": [
            "tsconvert_filter"
          ]
        }
      },
      "tokenizer": {
        "pinyin_tokenizer_1": {
          "type": "pinyin",
          "keep_separate_first_letter": false,
          "keep_full_pinyin": true,
          "keep_original": true,
          "limit_first_letter_length": 16,
          "lowercase": true,
          "remove_duplicated_term": true
        },
        "pinyin_tokenizer_2": {
          "type": "pinyin",
          "keep_separate_first_letter": true,
          "keep_full_pinyin&#