elasticsearch analysis ansj分词器的安装及使用

1. 修改pom文件配置

<elasticsearch.version>1.7.1</elasticsearch.version>
<dependency>
    <groupId>org.ansj</groupId>
    <artifactId>ansj_seg</artifactId>
    <classifier>min</classifier>
    <version>2.0.8</version>
    <scope>compile</scope>
</dependency>

2.编译插件

mvn assembly:assembly

3. 插件安装

elasticsearch-1.7.1\bin>plugin -u file:///C:\Users\Administrator\Desktop\elasticsearch-analysis-ansj\target\releases\elasticsearch-analysis-ansj-1.x.1-release.zip -i ansj

4. 配置ansj分词器

index:
  analysis:
    analyzer:
      index_ansj:
          type: ansj_index
      query_ansj:
          type: ansj_query
      ik:
          alias: [news_analyzer_ik,ik_analyzer]
          type: org.elasticsearch.index.analysis.IkAnalyzerProvider
      mmseg:
          alias: [news_analyzer, mmseg_analyzer]
          type: org.elasticsearch.index.analysis.MMsegAnalyzerProvider
        
index.analysis.analyzer.default.type : "ansj_index"

详细配置可参考elasticsearch.yml.example

5. 测试及使用

• 索引分词

http://127.0.0.1:9200/articles/_analyze?analyzer=ansj_index&text=我们是中国人

注：其中articles是索引名称，除articles外的所有请求url参数部分均为固定写法。analyzer=ansj_index指定索引分词器，text后为要索引的内容
输出：

{
  "tokens": [
    {
      "token": "我们",
      "start_offset": 0,
      "end_offset": 2,
      "type": "word",
      "position": 1
    },
    {
      "token": "是",
      "start_offset": 2,
      "end_offset": 3,
      "type": "word",
      "position": 2
    },
    {
      "token": "中国",
      "start_offset": 3,
      "end_offset": 5,
      "type": "word",
      "position": 3
    },
    {
      "token": "人",
      "start_offset": 5,
      "end_offset": 6,
      "type": "word",
      "position": 4
    }
  ]
}

• 查询分词

http://127.0.0.1:9200/articles/_analyze?analyzer=ansj_query&text=我们是中国人

注：其中articles是索引名称，除articles外的所有请求url参数部分均为固定写法。analyzer=ansj_query指定查询分词器，text后为要查询的内容
输出：

{
  "tokens": [
    {
      "token": "我们",
      "start_offset": 0,
      "end_offset": 2,
      "type": "word",
      "position": 1
    },
    {
      "token": "是",
      "start_offset": 2,
      "end_offset": 3,
      "type": "word",
      "position": 2
    },
    {
      "token": "中国",
      "start_offset": 3,
      "end_offset": 5,
      "type": "word",
      "position": 3
    },
    {
      "token": "人",
      "start_offset": 5,
      "end_offset": 6,
      "type": "word",
      "position": 4
    }
  ]
}