elasticsearch使用中文拼音混合搜索的痛点解决

在这里如果你想通过elasticsearch使用中文+拼音的分词,但是对于大家来说中文分词,拼音分词单独使用的时候会能合理运用,但是如果混合搜索就会出现一些问题,如果你有图中的这个样子的需求,那本文就是能帮助到你的了

我在这里是通过修改拼音插件分词的源码之后进行打包上传,然后安装之后就可以进行图中的搜索了.

es的版本:

elasticsearch 6.8.4

如果你正好是6.8.4,你就可以通过百度云盘下载下来直接使用了.

链接: https://pan.baidu.com/s/1GVtK6LeFuYWfbL5B6BotVQ 提取码: k4i8 复制这段内容后打开百度网盘手机App，操作更方便哦

如果不是这个版本之后,那么就是需要自己修改源码并编译了,我是通过这位大佬的博客中,你们也可以查看这个大佬的代码提示修改,我觉得已经很详细了:

[es 修改拼音分词器源码实现汉字/拼音/简拼混合搜索时同音字不匹配]

接下来就是在安装好了之后就是可以了,接下来就是贴代码了

@Test
public void setPyService() {
    String content = "武houq";
    HighlightBuilder highlightBuilder = new HighlightBuilder();

    /** 拼音字段 */
    SearchRequestBuilder searchRequestBuilderPy =
        client.prepareSearch("address").setTypes("address");
    BoolQueryBuilder boolQueryBuilderPy = QueryBuilders.boolQuery();

    highlightBuilder
        .field("region")
        .field("town")
        .field("villag")
        .field("road")
        .field("plotName");
    MatchPhraseQueryBuilder matchQueryBuilderRegionPy = QueryBuilders.matchPhraseQuery("region", content);
    MatchPhraseQueryBuilder matchQueryBuilderTownPy = QueryBuilders.matchPhraseQuery("town", content);
    MatchPhraseQueryBuilder matchQueryBuilderVillagPy = QueryBuilders.matchPhraseQuery("villag", content);
    MatchPhraseQueryBuilder termQueryBuilderRoad = QueryBuilders.matchPhraseQuery("road", content);
    MatchPhraseQueryBuilder matchQueryBuilderPnPy = QueryBuilders.matchPhraseQuery("plotName", content);


    boolQueryBuilderPy
        .should(matchQueryBuilderRegionPy)
        .should(matchQueryBuilderTownPy)
        .should(matchQueryBuilderVillagPy)
        .should( termQueryBuilderRoad)
        .should(matchQueryBuilderPnPy)
        ;
    searchRequestBuilderPy.setQuery(boolQueryBuilderPy).highlighter(highlightBuilder);
    SearchResponse searchResponse = searchRequestBuilderPy.execute().actionGet();
    for (SearchHit hit : searchResponse.getHits().getHits()) {
        System.out.println(hit.getSourceAsString());
    }
}

执行的结果:

{"id":"d06456165718469fbab524d513b1358c","region":"武侯区","town":"玉林街道办事处","villag":"黉门街社区","road":"黉门街29号","plotName":"武侯区教育局","dn":""}
{"id":"0c040fb2b5c94f89a5bda30995b26874","region":"武侯区","town":"浆洗街街道办事处","villag":"蜀汉社区","road":"蜀汉东街8号","plotName":"武侯区政府宿舍","dn":""}
{"id":"ed63768257234cca8c7fc5b1d35dd3ea","region":"武侯区","town":"双楠街道办事处","villag":"双楠街社区","road":"双楠街165号","plotName":"武侯区教师公寓","dn":"1幢"}
{"id":"f281711eaff140178167f810bc0de143","region":"武侯区","town":"双楠街道办事处","villag":"双楠街社区","road":"双楠街165号","plotName":"武侯区教师公寓","dn":"3幢"}
{"id":"ad18f51264fe4ef0ac01b5e918021834","region":"武侯区","town":"浆洗街街道办事处","villag":"蜀汉社区","road":"蜀汉东街8号","plotName":"武侯区政府宿舍","dn":"门卫室"}
{"id":"99b022cb06684dd9a247f8333fc189c1","region":"武侯区","town":"簇桥街道办事处","villag":"凉水井村","road":"","plotName":"武侯区鸿鑫皮鞋","dn":""}
{"id":"f05c75c071444e2b97476ef8e5564f57","region":"武侯区","town":"金花街道办事处","villag":"双凤村","road":"凤翔路","plotName":"武侯区兴宏医疗","dn":""}
{"id":"eb279f23eb63454e8a73691366f85570","region":"武侯区","town":"金花街道办事处","villag":"双凤村","road":"金兴中路197号","plotName":"武侯区伊人门诊","dn":""}
{"id":"8ea46ba4a5a940dfb7da6986f96a6a84","region":"武侯区","town":"浆洗街街道办事处","villag":"蜀汉社区","road":"蜀汉东街8号","plotName":"武侯区政府宿舍","dn":"2幢"}
{"id":"c6f3fa089e5c4c0e984dd07f4add77fc","region":"武侯区","town":"双楠街道办事处","villag":"双楠街社区","road":"双楠街165号","plotName":"武侯区教师公寓","dn":"2幢"}

这里可能有人配置了还是没有成功,我贴一下我的es的mapping和settiong设置吧
 这个setting可以直接使用,前提是你得有以下的插件:

elasticsearch-analysis-hanlp-6.8.4 中文分词
elasticsearch-analysis-ik-6.8.4 ik分词
elasticsearch-analysis-stconvert-6.8.4 繁体字转换
elasticsearch-analysis-pinyin-6.8.4 拼音分词

如果有以上几个就是可以直接使用

{
  "index": {
    "analysis": {
      "filter": {
        "edge_ngram_filter": {
          "type": "edge_ngram",
          "min_gram": 1,
          "max_gram": 50
        },
        "pinyin_simple_filter": {
          "type": "pinyin",
          "keep_separate_first_letter": false,
          "keep_full_pinyin": false,
          "keep_original": true,
          "keep_joined_full_pinyin": true,
          "limit_first_letter_length": 16,
          "lowercase": true,
          "remove_duplicated_term": true
        }
      },
      "char_filter": {
        "tsconvert": {
          "type": "stconvert",
          "convert_type": "t2s"
        }
      },
      "analyzer": {
        "pinyin_chinese_analyzer": {
          "tokenizer": "pinyin_tokenizer"
        },
        "pinyin_analyzer": {
          "tokenizer": "pinyin_chinese_tokenizer"
        },
        "ikSearchAnalyzer": {
          "type": "custom",
          "tokenizer": "ik_max_word",
          "char_filter": [
            "tsconvert"
          ]
        },
        "my_analyzer": {
          "type": "hanlp",
          "algorithm": "viterbi",
          "enableIndexMode": "true",
          "enableCustomDictionary": "true",
          "customDictionaryPath": "",
          "enableCustomDictionaryForcing": "true",
          "enableStopWord": "true",
          "stopWordDictionaryPath": "",
          "enableNumberQuantifierRecognize": "true",
          "enableNameRecognize": "true",
          "enableTranslatedNameRecognize": "true",
          "enableJapaneseNameRecognize": "true",
          "enableOrganizationRecognize": "true",
          "enablePlaceRecognize": "true",
          "enableTraditionalChineseMode": "false"
        },
        "pinyinSimpleIndexAnalyzer": {
          "tokenizer": "keyword",
          "filter": [
            "pinyin_simple_filter",
            "edge_ngram_filter",
            "lowercase",
            "edgeNGram",
            "standard"
          ]
        }
      },
      //这里按照我这样设置的话,最后返回的就是上面的键结果,当然如果你想自己定义,可以去拼音插件的github上自行添加
      "tokenizer": {
        "pinyin_chinese_tokenizer": {
          "type": "pinyin",
          "keep_first_letter": true,
          "keep_separate_first_letter": true,
          "keep_full_pinyin": true,
          "keep_original": true,
          "limit_first_letter_length": 50,
          "keep_separate_chinese": true,
          "lowercase": true,
          "keep_none_chinese_together": true
        },
        "pinyin_tokenizer": {
          "type": "pinyin",
          "keep_first_letter": true,
          "keep_separate_first_letter": true,
          "keep_full_pinyin": true,
          "keep_original": true,
          "limit_first_letter_length": 50,
          "keep_separate_chinese": true,
          "lowercase": true,
          "keep_none_chinese_together": true
        }
      }
    }
  }
}

{
  "address": {
    "_all": {
      "enabled": false
    },
    "properties": {
      "id": {
        "type": "keyword"
      },
      "region": {
        "type": "text",
        //按照这个方式修改即可,可以直接复制改名字就行了
        "search_analyzer": "pinyin_chinese_analyzer",
        "analyzer": "pinyin_analyzer"
      },
      "town": {
        "type": "text",
        "search_analyzer": "pinyin_chinese_analyzer",
        "analyzer": "pinyin_analyzer"
      },
      "villag": {
        "type": "text",
        "search_analyzer": "pinyin_chinese_analyzer",
        "analyzer": "pinyin_analyzer"
      },
      "road": {
        "type": "text",
        "search_analyzer": "pinyin_chinese_analyzer",
        "analyzer": "pinyin_analyzer"
      },
      "plotName": {
        "type": "text",
        "search_analyzer": "pinyin_chinese_analyzer",
        "analyzer": "pinyin_analyzer"
      },
      "dn": {
        "type": "text",
        "search_analyzer": "pinyin_chinese_analyzer",
        "analyzer": "pinyin_analyzer"
      }
    }
  }
}

 完毕