Elasticsearch之中文分词器插件es-ik的自定义热更新词库

不多说，直接上干货！

欢迎大家，关注微信扫码并加入我的4个微信公众号：   大数据躺过的坑      Java从入门到架构师      人工智能躺过的坑         Java全栈大联盟

 

     每天都有大量的学习视频资料和精彩技术文章推送... 人生不易，唯有努力。

 

     百家号 ：九月哥快讯               快手号：  jiuyuege

 

 

 

 

 

前提

Elasticsearch之中文分词器插件es-ik的自定义词库

　　先声明，热更新词库，需要用到，web项目和Tomcat。不会的，请移步

Eclipse下Maven新建项目、自动打依赖jar包（包含普通项目和Web项目）

Tomcat *的安装和运行（绿色版和安装版都适用）

Tomcat的配置文件详解

1： 部署 http 服务
在这使用 tomcat7 作为 web 容器， 先下载一个 tomcat7， 然后上传到某一台服务器上(192.168.80.10)。
再执行以下命令
　　tar -zxvf apache-tomcat-7.0.73.tar.gz
　　cd apache-tomcat-7.0.73/webapp/ROOT
vi hot.dic
　　测试

　　在这里，我是为了避免跟我的hadoop和spark集群里的端口冲突，将默认的tomcat8080端口，改为8081端口了。

在CentOS下安装tomcat并配置环境变量（改默认端口8080为8081）

　　如果，是3台tomcat集群的话，则对应，比如我的192.168.80.10位8081端口，192.168.80.11位8082端口，192.168.80.12位8083端口

验证一下这个文件是否可以正常访问

　　http://192.168.80.10:/zhoulshot.dic

2： 修改 ik 插件的配置文件
cd elasticsearch-2.4.3/plugins/ik/config
vi IKAnalyzer.cfg.xml
修改 key=remote_ext_dict 的 entry 中的内容
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE properties SYSTEM "http://java.sun.com/dtd/properties.dtd">
　　<properties>
　　<comment>IK Analyzer 扩展配置</comment>
　　<!--用户可以在这里配置自己的扩展字典 -->
　　<entry key="ext_dict">custom/mydict.dic;custom/single_word_low_freq.dic;custom/zhouls.dic</entry>
　　<!--用户可以在这里配置自己的扩展停止词字典-->
　　<entry key="ext_stopwords">custom/ext_stopword.dic</entry>
　　<!--用户可以在这里配置远程扩展字典 -->
　　<entry key="remote_ext_dict">http://192.168.80.10:8081/zhoulshot.dic</entry>
　　<!--用户可以在这里配置远程扩展停止词字典-->
　　<!-- <entry key="remote_ext_stopwords">words_location</entry> -->
</properties>

　　注意：（1）默认是words_location，我这里改为我自己的了。http://192.168.80.10:8081/zhoulshot.dic   （自定义词库）

　　　　  （2）默认是custom/mydict.dic;custom/single_word_low_freq.dic，我这里改为我自己的了。    （自定义热更新词库）   custom/mydict.dic;custom/single_word_low_freq.dic;custom/zhouls.dic

3： 验证
　　重启 es， 会看到如下日志信息， 说明远程的词典加载成功了。

执行下面命令查看分词效果
　　curl 'http://192.168.80.10:9200/zhouls/_analyze?analyzer=ik_max_word&pretty=true' -d '{"text":"桂林山水"}'
正常情况下桂林山水会分为多个词语， 但是我们希望 es 把[桂林山水]作为一个完整的词， 又不希望重启 es。
这样就需要修改前面的 zhoulshot.dic 文件， 增加一个词语[桂林山水]

vi hot.dic
　　桂林山水
文件保存之后， 查看 es 的日志会看到如下日志信息

再执行下面命令查看分词效果
　　curl 'http://192.168.80.10:9200/zhouls/_analyze?analyzer=ik_max_word&pretty=true' -d '{"text":"桂林山水"}'
到这为止， 可以实现动态添加自定义词库实现词库热更新。
==============================================================================
　　注意： 默认情况下， 最多一分钟之内就可以识别到新增的词语。

查看 es-ik 插件的源码可以发现

1： 部署 http 服务

　　第一步：下载tomcat压缩包

http://archive.apache.org/dist/tomcat/tomcat-7/v7.0.73/bin/

　

　

　　第二步：上传tomcat压缩包

[hadoop@HadoopMaster app]$ ll
total 3092
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 22 06:05 elasticsearch-2.4.3
-rw-r--r--. 1 hadoop hadoop 908862 Jan 10 11:38 elasticsearch-head-master.zip
-rw-r--r--. 1 hadoop hadoop 2228252 Jan 10 11:38 elasticsearch-kopf-master.zip
drwxr-xr-x. 10 hadoop hadoop 4096 Oct 31 17:15 hadoop-2.6.0
drwxr-xr-x. 15 hadoop hadoop 4096 Nov 14 2014 hadoop-2.6.0-src
drwxrwxr-x. 8 hadoop hadoop 4096 Nov 2 18:20 hbase-1.2.3
drwxr-xr-x. 8 hadoop hadoop 4096 Apr 11 2015 jdk1.7.0_79
drwxr-xr-x. 10 hadoop hadoop 4096 Nov 1 23:39 zookeeper-3.4.6
[hadoop@HadoopMaster app]$ rz

[hadoop@HadoopMaster app]$ ll
total 11824
-rw-r--r--. 1 hadoop hadoop 8938514 Feb 25 11:10 apache-tomcat-7.0.73.tar.gz
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 22 06:05 elasticsearch-2.4.3
-rw-r--r--. 1 hadoop hadoop 908862 Jan 10 11:38 elasticsearch-head-master.zip
-rw-r--r--. 1 hadoop hadoop 2228252 Jan 10 11:38 elasticsearch-kopf-master.zip
drwxr-xr-x. 10 hadoop hadoop 4096 Oct 31 17:15 hadoop-2.6.0
drwxr-xr-x. 15 hadoop hadoop 4096 Nov 14 2014 hadoop-2.6.0-src
drwxrwxr-x. 8 hadoop hadoop 4096 Nov 2 18:20 hbase-1.2.3
drwxr-xr-x. 8 hadoop hadoop 4096 Apr 11 2015 jdk1.7.0_79
drwxr-xr-x. 10 hadoop hadoop 4096 Nov 1 23:39 zookeeper-3.4.6
[hadoop@HadoopMaster app]$

　　第三步:解压缩

[hadoop@HadoopMaster app]$ tar -zxvf apache-tomcat-7.0.73.tar.gz

　　第四步：删除压缩包

[hadoop@HadoopMaster app]$ ll
total 11828
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 25 19:18 apache-tomcat-7.0.73
-rw-r--r--. 1 hadoop hadoop 8938514 Feb 25 11:10 apache-tomcat-7.0.73.tar.gz
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 22 06:05 elasticsearch-2.4.3
-rw-r--r--. 1 hadoop hadoop 908862 Jan 10 11:38 elasticsearch-head-master.zip
-rw-r--r--. 1 hadoop hadoop 2228252 Jan 10 11:38 elasticsearch-kopf-master.zip
drwxr-xr-x. 10 hadoop hadoop 4096 Oct 31 17:15 hadoop-2.6.0
drwxr-xr-x. 15 hadoop hadoop 4096 Nov 14 2014 hadoop-2.6.0-src
drwxrwxr-x. 8 hadoop hadoop 4096 Nov 2 18:20 hbase-1.2.3
drwxr-xr-x. 8 hadoop hadoop 4096 Apr 11 2015 jdk1.7.0_79
drwxr-xr-x. 10 hadoop hadoop 4096 Nov 1 23:39 zookeeper-3.4.6
[hadoop@HadoopMaster app]$ rm apache-tomcat-7.0.73.tar.gz
[hadoop@HadoopMaster app]$ ll
total 3096
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 25 19:18 apache-tomcat-7.0.73
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 22 06:05 elasticsearch-2.4.3
-rw-r--r--. 1 hadoop hadoop 908862 Jan 10 11:38 elasticsearch-head-master.zip
-rw-r--r--. 1 hadoop hadoop 2228252 Jan 10 11:38 elasticsearch-kopf-master.zip
drwxr-xr-x. 10 hadoop hadoop 4096 Oct 31 17:15 hadoop-2.6.0
drwxr-xr-x. 15 hadoop hadoop 4096 Nov 14 2014 hadoop-2.6.0-src
drwxrwxr-x. 8 hadoop hadoop 4096 Nov 2 18:20 hbase-1.2.3
drwxr-xr-x. 8 hadoop hadoop 4096 Apr 11 2015 jdk1.7.0_79
drwxr-xr-x. 10 hadoop hadoop 4096 Nov 1 23:39 zookeeper-3.4.6
[hadoop@HadoopMaster app]$

　　第五步：重命名tomcat安装目录

[hadoop@HadoopMaster app]$ ll
total 3096
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 25 19:18 apache-tomcat-7.0.73
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 22 06:05 elasticsearch-2.4.3
-rw-r--r--. 1 hadoop hadoop 908862 Jan 10 11:38 elasticsearch-head-master.zip
-rw-r--r--. 1 hadoop hadoop 2228252 Jan 10 11:38 elasticsearch-kopf-master.zip
drwxr-xr-x. 10 hadoop hadoop 4096 Oct 31 17:15 hadoop-2.6.0
drwxr-xr-x. 15 hadoop hadoop 4096 Nov 14 2014 hadoop-2.6.0-src
drwxrwxr-x. 8 hadoop hadoop 4096 Nov 2 18:20 hbase-1.2.3
drwxr-xr-x. 8 hadoop hadoop 4096 Apr 11 2015 jdk1.7.0_79
drwxr-xr-x. 10 hadoop hadoop 4096 Nov 1 23:39 zookeeper-3.4.6
[hadoop@HadoopMaster app]$ mv apache-tomcat-7.0.73 tomcat-7.0.73
[hadoop@HadoopMaster app]$ ll
total 3096
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 22 06:05 elasticsearch-2.4.3
-rw-r--r--. 1 hadoop hadoop 908862 Jan 10 11:38 elasticsearch-head-master.zip
-rw-r--r--. 1 hadoop hadoop 2228252 Jan 10 11:38 elasticsearch-kopf-master.zip
drwxr-xr-x. 10 hadoop hadoop 4096 Oct 31 17:15 hadoop-2.6.0
drwxr-xr-x. 15 hadoop hadoop 4096 Nov 14 2014 hadoop-2.6.0-src
drwxrwxr-x. 8 hadoop hadoop 4096 Nov 2 18:20 hbase-1.2.3
drwxr-xr-x. 8 hadoop hadoop 4096 Apr 11 2015 jdk1.7.0_79
drwxrwxr-x. 9 hadoop hadoop 4096 Feb 25 19:18 tomcat-7.0.73
drwxr-xr-x. 10 hadoop hadoop 4096 Nov 1 23:39 zookeeper-3.4.6
[hadoop@HadoopMaster app]$

　　第六步：进入tomcat安装目录，并初步认识下

[hadoop@HadoopMaster app]$ cd tomcat-7.0.73/
[hadoop@HadoopMaster tomcat-7.0.73]$ ll
total 116
drwxr-xr-x. 2 hadoop hadoop 4096 Feb 25 19:18 bin
drwxr-xr-x. 2 hadoop hadoop 4096 Nov 8 05:30 conf
drwxr-xr-x. 2 hadoop hadoop 4096 Feb 25 19:18 lib
-rw-r--r--. 1 hadoop hadoop 56846 Nov 8 05:30 LICENSE
drwxr-xr-x. 2 hadoop hadoop 4096 Nov 8 05:27 logs
-rw-r--r--. 1 hadoop hadoop 1239 Nov 8 05:30 NOTICE
-rw-r--r--. 1 hadoop hadoop 8965 Nov 8 05:30 RELEASE-NOTES
-rw-r--r--. 1 hadoop hadoop 16195 Nov 8 05:30 RUNNING.txt
drwxr-xr-x. 2 hadoop hadoop 4096 Feb 25 19:18 temp
drwxr-xr-x. 7 hadoop hadoop 4096 Nov 8 05:29 webapps
drwxr-xr-x. 2 hadoop hadoop 4096 Nov 8 05:27 work
[hadoop@HadoopMaster tomcat-7.0.73]$

　　在这里，需要，先事先好，在linux下安装好tomcat。并配置好环境变量。不会的，请移步

在CentOS下安装tomcat并配置环境变量

　　第七步：进入webapps/ROOT目录下

[hadoop@HadoopMaster tomcat-7.0.73]$ cd webapps/
[hadoop@HadoopMaster webapps]$ pwd
/home/hadoop/app/tomcat-7.0.73/webapps
[hadoop@HadoopMaster webapps]$ ll
total 20
drwxr-xr-x. 14 hadoop hadoop 4096 Feb 25 19:18 docs
drwxr-xr-x. 7 hadoop hadoop 4096 Feb 25 19:18 examples
drwxr-xr-x. 5 hadoop hadoop 4096 Feb 25 19:18 host-manager
drwxr-xr-x. 5 hadoop hadoop 4096 Feb 25 19:18 manager
drwxr-xr-x. 3 hadoop hadoop 4096 Feb 25 19:18 ROOT
[hadoop@HadoopMaster webapps]$ cd ROOT/
[hadoop@HadoopMaster ROOT]$ pwd
/home/hadoop/app/tomcat-7.0.73/webapps/ROOT
[hadoop@HadoopMaster ROOT]$ ll
total 196
-rw-r--r--. 1 hadoop hadoop 17811 Nov 8 05:29 asf-logo.png
-rw-r--r--. 1 hadoop hadoop 5866 Nov 8 05:29 asf-logo-wide.gif
-rw-r--r--. 1 hadoop hadoop 713 Nov 8 05:29 bg-button.png
-rw-r--r--. 1 hadoop hadoop 1918 Nov 8 05:29 bg-middle.png
-rw-r--r--. 1 hadoop hadoop 1392 Nov 8 05:29 bg-nav-item.png
-rw-r--r--. 1 hadoop hadoop 1401 Nov 8 05:29 bg-nav.png
-rw-r--r--. 1 hadoop hadoop 3103 Nov 8 05:29 bg-upper.png
-rw-r--r--. 1 hadoop hadoop 3376 Nov 8 05:30 build.xml
-rw-r--r--. 1 hadoop hadoop 21630 Nov 8 05:29 favicon.ico
-rw-r--r--. 1 hadoop hadoop 12186 Nov 8 05:30 index.jsp
-rw-r--r--. 1 hadoop hadoop 8965 Nov 8 05:30 RELEASE-NOTES.txt
-rw-r--r--. 1 hadoop hadoop 5576 Nov 8 05:30 tomcat.css
-rw-r--r--. 1 hadoop hadoop 2066 Nov 8 05:29 tomcat.gif
-rw-r--r--. 1 hadoop hadoop 5103 Nov 8 05:29 tomcat.png
-rw-r--r--. 1 hadoop hadoop 2376 Nov 8 05:29 tomcat-power.gif
-rw-r--r--. 1 hadoop hadoop 67198 Nov 8 05:30 tomcat.svg
drwxr-xr-x. 2 hadoop hadoop 4096 Feb 25 19:18 WEB-INF
[hadoop@HadoopMaster ROOT]$

　　第八步：新建，自定义的热更新词库。如，我这里，是，zhoulshot.dic

[hadoop@HadoopMaster ROOT]$ pwd
/home/hadoop/app/tomcat-7.0.73/webapps/ROOT
[hadoop@HadoopMaster ROOT]$ vim zhoulshot.dic
[hadoop@HadoopMaster ROOT]$ cat zhoulshot.dic
好记性不如烂笔头感叹号博客园热更新词
[hadoop@HadoopMaster ROOT]$ ll
total 200
-rw-r--r--. 1 hadoop hadoop 17811 Nov 8 05:29 asf-logo.png
-rw-r--r--. 1 hadoop hadoop 5866 Nov 8 05:29 asf-logo-wide.gif
-rw-r--r--. 1 hadoop hadoop 713 Nov 8 05:29 bg-button.png
-rw-r--r--. 1 hadoop hadoop 1918 Nov 8 05:29 bg-middle.png
-rw-r--r--. 1 hadoop hadoop 1392 Nov 8 05:29 bg-nav-item.png
-rw-r--r--. 1 hadoop hadoop 1401 Nov 8 05:29 bg-nav.png
-rw-r--r--. 1 hadoop hadoop 3103 Nov 8 05:29 bg-upper.png
-rw-r--r--. 1 hadoop hadoop 3376 Nov 8 05:30 build.xml
-rw-r--r--. 1 hadoop hadoop 21630 Nov 8 05:29 favicon.ico
-rw-r--r--. 1 hadoop hadoop 12186 Nov 8 05:30 index.jsp
-rw-r--r--. 1 hadoop hadoop 8965 Nov 8 05:30 RELEASE-NOTES.txt
-rw-r--r--. 1 hadoop hadoop 5576 Nov 8 05:30 tomcat.css
-rw-r--r--. 1 hadoop hadoop 2066 Nov 8 05:29 tomcat.gif
-rw-r--r--. 1 hadoop hadoop 5103 Nov 8 05:29 tomcat.png
-rw-r--r--. 1 hadoop hadoop 2376 Nov 8 05:29 tomcat-power.gif
-rw-r--r--. 1 hadoop hadoop 67198 Nov 8 05:30 tomcat.svg
drwxr-xr-x. 2 hadoop hadoop 4096 Feb 25 19:18 WEB-INF
-rw-rw-r--. 1 hadoop hadoop 55 Feb 25 19:44 zhoulshot.dic
[hadoop@HadoopMaster ROOT]$

　　第九步：验证一下这个zhoulshot热更新词文件是否可以正常访问　

　　

2： 修改 ik 插件的配置文件

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE properties SYSTEM "http://java.sun.com/dtd/properties.dtd">
<properties>
<comment>IK Analyzer 扩展配置</comment>
<!--用户可以在这里配置自己的扩展字典 -->
<entry key="ext_dict">custom/mydict.dic;custom/single_word_low_freq.dic;custom/zhouls.dic</entry>
<!--用户可以在这里配置自己的扩展停止词字典-->
<entry key="ext_stopwords">custom/ext_stopword.dic</entry>
<!--用户可以在这里配置远程扩展字典 -->
<!-- <entry key="remote_ext_dict">words_location</entry> -->
<!--用户可以在这里配置远程扩展停止词字典-->
<!-- <entry key="remote_ext_stopwords">words_location</entry> -->
</properties>

修改为

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE properties SYSTEM "http://java.sun.com/dtd/properties.dtd">
<properties>
<comment>IK Analyzer 扩展配置</comment>
<!--用户可以在这里配置自己的扩展字典 -->
<entry key="ext_dict">custom/mydict.dic;custom/single_word_low_freq.dic;custom/zhouls.dic</entry>
<!--用户可以在这里配置自己的扩展停止词字典-->
<entry key="ext_stopwords">custom/ext_stopword.dic</entry>
<!--用户可以在这里配置远程扩展字典 -->
<entry key="remote_ext_dict">http://192.168.80.10:8081/zhoulshot.dic</entry>
<!--用户可以在这里配置远程扩展停止词字典-->
<!-- <entry key="remote_ext_stopwords">words_location</entry> -->
</properties>

3： 验证
　　重启 es， 会看到如下日志信息， 说明远程的词典加载成功了。

我这里，为了更好地看出效果，在es的安装目录下，bin/elasticsearch这种方式来启动。

[hadoop@HadoopMaster elasticsearch-2.4.3]$ jps
2115 Jps
1979 Bootstrap
[hadoop@HadoopMaster elasticsearch-2.4.3]$ bin/elasticsearch
[2017-02-26 18:02:00,383][WARN ][bootstrap ] unable to install syscall filter: seccomp unavailable: requires kernel 3.5+ with CONFIG_SECCOMP and CONFIG_SECCOMP_FILTER compiled in
[2017-02-26 18:02:01,915][INFO ][node ] [Meteor Man] version[2.4.3], pid[2126], build[d38a34e/2016-12-07T16:28:56Z]
[2017-02-26 18:02:01,915][INFO ][node ] [Meteor Man] initializing ...
[2017-02-26 18:02:06,929][INFO ][plugins ] [Meteor Man] modules [lang-groovy, reindex, lang-expression], plugins [analysis-ik, kopf, head], sites [kopf, head]
[2017-02-26 18:02:07,141][INFO ][env ] [Meteor Man] using [1] data paths, mounts [[/home (/dev/sda5)]], net usable_space [23.4gb], net total_space [26.1gb], spins? [possibly], types [ext4]
[2017-02-26 18:02:07,141][INFO ][env ] [Meteor Man] heap size [1015.6mb], compressed ordinary object pointers [true]
[2017-02-26 18:02:07,142][WARN ][env ] [Meteor Man] max file descriptors [4096] for elasticsearch process likely too low, consider increasing to at least [65536]
[2017-02-26 18:02:12,726][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/config/analysis-ik/IKAnalyzer.cfg.xml
[2017-02-26 18:02:12,728][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/plugins/ik/config/IKAnalyzer.cfg.xml
[2017-02-26 18:02:13,763][INFO ][ik-analyzer ] [Dict Loading] custom/mydict.dic
[2017-02-26 18:02:13,799][INFO ][ik-analyzer ] [Dict Loading] custom/single_word_low_freq.dic
[2017-02-26 18:02:13,816][INFO ][ik-analyzer ] [Dict Loading] custom/zhouls.dic
[2017-02-26 18:02:13,821][INFO ][ik-analyzer ] [Dict Loading] http://192.168.80.10:8081/zhoulshot.dic
[2017-02-26 18:02:15,328][INFO ][ik-analyzer ] 好记性不如烂笔头感叹号博客园热更新词
[2017-02-26 18:02:15,394][INFO ][ik-analyzer ] [Dict Loading] custom/ext_stopword.dic
[2017-02-26 18:02:16,766][INFO ][node ] [Meteor Man] initialized
[2017-02-26 18:02:16,766][INFO ][node ] [Meteor Man] starting ...
[2017-02-26 18:02:18,221][INFO ][transport ] [Meteor Man] publish_address {192.168.80.10:9300}, bound_addresses {[::]:9300}
[2017-02-26 18:02:18,257][INFO ][discovery ] [Meteor Man] elasticsearch/EkiwUFTnTZO4PzCVERAukw
[2017-02-26 18:02:21,460][INFO ][cluster.service ] [Meteor Man] new_master {Meteor Man}{EkiwUFTnTZO4PzCVERAukw}{192.168.80.10}{192.168.80.10:9300}, reason: zen-disco-join(elected_as_master, [0] joins received)
[2017-02-26 18:02:21,538][INFO ][http ] [Meteor Man] publish_address {192.168.80.10:9200}, bound_addresses {[::]:9200}
[2017-02-26 18:02:21,542][INFO ][node ] [Meteor Man] started
[2017-02-26 18:02:22,376][INFO ][gateway ] [Meteor Man] recovered [1] indices into cluster_state
[2017-02-26 18:02:25,460][INFO ][ik-analyzer ] 重新加载词典...
[2017-02-26 18:02:25,462][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/config/analysis-ik/IKAnalyzer.cfg.xml
[2017-02-26 18:02:25,468][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/plugins/ik/config/IKAnalyzer.cfg.xml

[2017-02-26 18:02:27,090][INFO ][ik-analyzer ] [Dict Loading] custom/mydict.dic
[2017-02-26 18:02:27,092][INFO ][ik-analyzer ] [Dict Loading] custom/single_word_low_freq.dic
[2017-02-26 18:02:27,097][INFO ][ik-analyzer ] [Dict Loading] custom/zhouls.dic
[2017-02-26 18:02:27,098][INFO ][ik-analyzer ] [Dict Loading] http://192.168.80.10:8081/zhoulshot.dic
[2017-02-26 18:02:27,134][INFO ][ik-analyzer ] 好记性不如烂笔头感叹号博客园热更新词
[2017-02-26 18:02:27,138][INFO ][ik-analyzer ] [Dict Loading] custom/ext_stopword.dic
[2017-02-26 18:02:27,140][INFO ][ik-analyzer ] 重新加载词典完毕...

　　执行下面命令查看分词效果

[hadoop@HadoopMaster elasticsearch-2.4.3]$ jps
2283 Jps
2195 Elasticsearch
1979 Bootstrap
[hadoop@HadoopMaster elasticsearch-2.4.3]$ curl 'http://192.168.80.10:9200/zhouls/_analyze?analyzer=ik_max_word&pretty=true' -d '{"text":"桂林不雾霾"}'
{
"tokens" : [ {
"token" : "桂林",
"start_offset" : 0,
"end_offset" : 2,
"type" : "CN_WORD",
"position" : 0
}, {
"token" : "桂",
"start_offset" : 0,
"end_offset" : 1,
"type" : "CN_WORD",
"position" : 1
}, {
"token" : "林",
"start_offset" : 1,
"end_offset" : 2,
"type" : "CN_CHAR",
"position" : 2
}, {
"token" : "不",
"start_offset" : 2,
"end_offset" : 3,
"type" : "CN_CHAR",
"position" : 3
}, {
"token" : "雾",
"start_offset" : 3,

"end_offset" : 4,
"type" : "CN_WORD",
"position" : 4
}, {
"token" : "霾",
"start_offset" : 4,
"end_offset" : 5,
"type" : "CN_WORD",
"position" : 5
} ]
}
[hadoop@HadoopMaster elasticsearch-2.4.3]$

　　正常情况下桂林不雾霾会分为多个词语， 但是我们希望 es 把[桂林不雾霾]作为一个完整的词， 又不希望重启 es。
　　这样就需要修改前面的 zhoulshot.dic 文件， 增加一个词语[桂林不雾霾]
vim zhoulshot.dic

[hadoop@HadoopMaster ROOT]$ pwd
/home/hadoop/app/tomcat-7.0.73/webapps/ROOT
[hadoop@HadoopMaster ROOT]$ ll
total 200
-rw-r--r--. 1 hadoop hadoop 17811 Nov 8 05:29 asf-logo.png
-rw-r--r--. 1 hadoop hadoop 5866 Nov 8 05:29 asf-logo-wide.gif
-rw-r--r--. 1 hadoop hadoop 713 Nov 8 05:29 bg-button.png
-rw-r--r--. 1 hadoop hadoop 1918 Nov 8 05:29 bg-middle.png
-rw-r--r--. 1 hadoop hadoop 1392 Nov 8 05:29 bg-nav-item.png
-rw-r--r--. 1 hadoop hadoop 1401 Nov 8 05:29 bg-nav.png
-rw-r--r--. 1 hadoop hadoop 3103 Nov 8 05:29 bg-upper.png
-rw-r--r--. 1 hadoop hadoop 3376 Nov 8 05:30 build.xml
-rw-r--r--. 1 hadoop hadoop 21630 Nov 8 05:29 favicon.ico
-rw-r--r--. 1 hadoop hadoop 12186 Nov 8 05:30 index.jsp
-rw-r--r--. 1 hadoop hadoop 8965 Nov 8 05:30 RELEASE-NOTES.txt
-rw-r--r--. 1 hadoop hadoop 5576 Nov 8 05:30 tomcat.css
-rw-r--r--. 1 hadoop hadoop 2066 Nov 8 05:29 tomcat.gif
-rw-r--r--. 1 hadoop hadoop 5103 Nov 8 05:29 tomcat.png
-rw-r--r--. 1 hadoop hadoop 2376 Nov 8 05:29 tomcat-power.gif
-rw-r--r--. 1 hadoop hadoop 67198 Nov 8 05:30 tomcat.svg
drwxr-xr-x. 2 hadoop hadoop 4096 Feb 25 19:18 WEB-INF
-rw-rw-r--. 1 hadoop hadoop 55 Feb 25 19:44 zhoulshot.dic
[hadoop@HadoopMaster ROOT]$ vim zhoulshot.dic

[hadoop@HadoopMaster ROOT]$ cat zhoulshot.dic
好记性不如烂笔头感叹号博客园热更新词
桂林不雾霾
[hadoop@HadoopMaster ROOT]$

　　文件保存之后， 重启es，查看 es 的日志会看到如下日志信息

[hadoop@HadoopMaster elasticsearch-2.4.3]$ jps
2353 Jps
1979 Bootstrap
[hadoop@HadoopMaster elasticsearch-2.4.3]$ bin/elasticsearch
[2017-02-26 18:15:11,538][WARN ][bootstrap ] unable to install syscall filter: seccomp unavailable: requires kernel 3.5+ with CONFIG_SECCOMP and CONFIG_SECCOMP_FILTER compiled in
[2017-02-26 18:15:12,476][INFO ][node ] [Andromeda] version[2.4.3], pid[2363], build[d38a34e/2016-12-07T16:28:56Z]
[2017-02-26 18:15:12,476][INFO ][node ] [Andromeda] initializing ...
[2017-02-26 18:15:14,047][INFO ][plugins ] [Andromeda] modules [lang-groovy, reindex, lang-expression], plugins [analysis-ik, kopf, head], sites [kopf, head]
[2017-02-26 18:15:14,093][INFO ][env ] [Andromeda] using [1] data paths, mounts [[/home (/dev/sda5)]], net usable_space [23.4gb], net total_space [26.1gb], spins? [possibly], types [ext4]
[2017-02-26 18:15:14,094][INFO ][env ] [Andromeda] heap size [1015.6mb], compressed ordinary object pointers [true]
[2017-02-26 18:15:14,094][WARN ][env ] [Andromeda] max file descriptors [4096] for elasticsearch process likely too low, consider increasing to at least [65536]
[2017-02-26 18:15:17,940][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/config/analysis-ik/IKAnalyzer.cfg.xml
[2017-02-26 18:15:17,942][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/plugins/ik/config/IKAnalyzer.cfg.xml
[2017-02-26 18:15:18,804][INFO ][ik-analyzer ] [Dict Loading] custom/mydict.dic
[2017-02-26 18:15:18,805][INFO ][ik-analyzer ] [Dict Loading] custom/single_word_low_freq.dic
[2017-02-26 18:15:18,809][INFO ][ik-analyzer ] [Dict Loading] custom/zhouls.dic
[2017-02-26 18:15:18,809][INFO ][ik-analyzer ] [Dict Loading] http://192.168.80.10:8081/zhoulshot.dic
[2017-02-26 18:15:19,962][INFO ][ik-analyzer ] 好记性不如烂笔头感叹号博客园热更新词
[2017-02-26 18:15:19,964][INFO ][ik-analyzer ] 桂林不雾霾
[2017-02-26 18:15:19,979][INFO ][ik-analyzer ] [Dict Loading] custom/ext_stopword.dic
[2017-02-26 18:15:21,455][INFO ][node ] [Andromeda] initialized
[2017-02-26 18:15:21,455][INFO ][node ] [Andromeda] starting ...
[2017-02-26 18:15:21,557][INFO ][transport ] [Andromeda] publish_address {192.168.80.10:9300}, bound_addresses {[::]:9300}
[2017-02-26 18:15:21,565][INFO ][discovery ] [Andromeda] elasticsearch/U318aiv6RTi3dKHaCIRHWw
[2017-02-26 18:15:24,761][INFO ][cluster.service ] [Andromeda] new_master {Andromeda}{U318aiv6RTi3dKHaCIRHWw}{192.168.80.10}{192.168.80.10:9300}, reason: zen-disco-join(elected_as_master, [0] joins received)
[2017-02-26 18:15:24,914][INFO ][http ] [Andromeda] publish_address {192.168.80.10:9200}, bound_addresses {[::]:9200}
[2017-02-26 18:15:24,920][INFO ][node ] [Andromeda] started
[2017-02-26 18:15:25,640][INFO ][gateway ] [Andromeda] recovered [1] indices into cluster_state
[2017-02-26 18:15:30,044][INFO ][ik-analyzer ] 重新加载词典...
[2017-02-26 18:15:30,048][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/config/analysis-ik/IKAnalyzer.cfg.xml

[2017-02-26 18:15:30,051][INFO ][ik-analyzer ] try load config from /home/hadoop/app/elasticsearch-2.4.3/plugins/ik/config/IKAnalyzer.cfg.xml
[2017-02-26 18:15:31,617][INFO ][ik-analyzer ] [Dict Loading] custom/mydict.dic
[2017-02-26 18:15:31,618][INFO ][ik-analyzer ] [Dict Loading] custom/single_word_low_freq.dic
[2017-02-26 18:15:31,625][INFO ][ik-analyzer ] [Dict Loading] custom/zhouls.dic
[2017-02-26 18:15:31,627][INFO ][ik-analyzer ] [Dict Loading] http://192.168.80.10:8081/zhoulshot.dic
[2017-02-26 18:15:31,664][INFO ][ik-analyzer ] 好记性不如烂笔头感叹号博客园热更新词
[2017-02-26 18:15:31,665][INFO ][ik-analyzer ] 桂林不雾霾
[2017-02-26 18:15:31,666][INFO ][ik-analyzer ] [Dict Loading] custom/ext_stopword.dic
[2017-02-26 18:15:31,668][INFO ][ik-analyzer ] 重新加载词典完毕...

　　直接，ctrl + c，终止es进程，再bin/elasticsearch -d

　　再执行下面命令查看分词效果

[hadoop@HadoopMaster elasticsearch-2.4.3]$ curl 'http://192.168.80.10:9200/zhouls/_analyze?analyzer=ik_max_word&pretty=true' -d '{"text":"桂林不雾霾"}'
{
"tokens" : [ {
"token" : "桂林不雾霾",
"start_offset" : 0,
"end_offset" : 5,
"type" : "CN_WORD",
"position" : 0
}, {
"token" : "桂林",
"start_offset" : 0,
"end_offset" : 2,
"type" : "CN_WORD",
"position" : 1
}, {
"token" : "桂",
"start_offset" : 0,
"end_offset" : 1,
"type" : "CN_WORD",
"position" : 2
}, {
"token" : "林",
"start_offset" : 1,
"end_offset" : 2,
"type" : "CN_CHAR",
"position" : 3
}, {
"token" : "不",
"start_offset" : 2,
"end_offset" : 3,
"type" : "CN_CHAR",
"position" : 4
}, {

"token" : "雾",
"start_offset" : 3,
"end_offset" : 4,
"type" : "CN_WORD",
"position" : 5
}, {
"token" : "霾",
"start_offset" : 4,
"end_offset" : 5,
"type" : "CN_WORD",
"position" : 6
} ]
}
[hadoop@HadoopMaster elasticsearch-2.4.3]$

　　到这为止， 可以实现动态添加自定义词库实现词库热更新!
==============================================================================
注意： 默认情况下， 最多一分钟之内就可以识别到新增的词语。

　　完毕！

欢迎大家，加入我的4个微信公众号：    大数据躺过的坑     Java从入门到架构师    人工智能躺过的坑     Java全栈大联盟    

 

 

 

同时，大家可以关注我的个人博客：

   http://www.cnblogs.com/zlslch/   和     http://www.cnblogs.com/lchzls/      http://www.cnblogs.com/sunnyDream/   

   详情请见：http://www.cnblogs.com/zlslch/p/7473861.html

　　人生苦短，我愿分享。本公众号将秉持活到老学到老学习无休止的交流分享开源精神，汇聚于互联网和个人学习工作的精华干货知识，一切来于互联网，反馈回互联网。
　　目前研究领域：大数据、机器学习、深度学习、人工智能、数据挖掘、数据分析。 语言涉及：Java、Scala、Python、Shell、Linux等 。同时还涉及平常所使用的手机、电脑和互联网上的使用技巧、问题和实用软件。 只要你一直关注和呆在群里，每天必须有收获

对应本平台的讨论和答疑QQ群：大数据和人工智能躺过的坑（总群）（161156071） 

打开百度App，扫码，精彩文章每天更新！欢迎关注我的百家号： 九月哥快讯