一.运行环境
1.操作系统Centos64
2.ElasticSearch版本5.6.3
3.Kibana版本5.6.3
二.ES安装
1.下载tar包
https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-5.6.3.tar.gz
2.解压tar包
tar -zxvf elasticsearch-5.6.3.tar.gz -C /home/elasticsearch-5.6.3
3.修改elasticsearch.yml配置文件
vi /home/elasticsearch-5.6.3/config/elasticsearch.yml
cluster.name: nlp
node.name: search
network.host: 0.0.0.0
http.port: 9200
4.elasticsearch不能以root用户启动,需要新建用户组和用户
groupadd nlp
useradd nlp -g nlp
chown -R nlp:nlp /home/elasticsearch-5.6.3
5.启动es
su nlp
./bin/elasticsearch -d
6.访问127.0.0.1:9200,看到如下返回表示启动正常
{ "name" : "search", "cluster_name" : "nlp", "cluster_uuid" : "aF1AO6NfRBeCF1k9Xlb8hA", "version" : { "number" : "5.6.3", "build_hash" : "1a2f265", "build_date" : "2017-10-06T20:33:39.012Z", "build_snapshot" : false, "lucene_version" : "6.6.1" }, "tagline" : "You Know, for Search" }
三.kibana安装
1.返回到root用户
2.下载kibana安装包,由于官网下载速度较慢,采用华为镜像安装。注意kibana安装包版本号要与elasticsearch版本号一致。
https://mirrors.huaweicloud.com/kibana/
kibana-5.6.3-linux-x86_64.tar.gz
3.解压压缩包到/home/kibana-5.6.3目录
tar -zxvf kibana-5.6.3-linux-x86_64.tar.gz -C /home/kibana-5.6.3
4.修改配置文件
vi /home/kibana-5.6.3/config/kibana.yml
server.port: 5601
server.host: "0.0.0.0"
elasticsearch.url: http://127.0.0.1:9200
5.将kibana目录分配给nlp用户
chown -R nlp:nlp /home/kibana-5.6.3
6.转到nlp用户启动kibana
su nlp
nohup ./kibana &
7.访问http://127.0.0.1:5601/app/kibana,展示如下界面表示安装成功
四.hanlp插件安装
1.下载hanlp插件
链接: https://pan.baidu.com/s/1pMdfzkB 密码: cdc4
2.将hanlp插件解压到/home/elasticsearch-5.6.3/plugins目录下
unzip elasticsearch-hanlp-5.6.3.zip -d /home/elasticsearch-5.6.3/plugins
3.将hanlp目录分配给nlp用户
chown -R nlp:nlp /home/elasticsearch-5.6.3/plugins/hanlp
4.删除hanlp目录下log4j-api-2.9.1.jar,防止jar包冲突
5.下载hanlp数据包
链接: https://pan.baidu.com/s/1smsAxch 密码: w4i7
解压之后,修改/
elasticsearch-5.6.3/
plugins/
hanlp目录下的hanlp.properties文件,修改根路径root=/home/hanlp/
为数据包所在目录
6.修改plugin-security.policy文件,在末尾添加
-Djava.security.policy=/home/elasticsearch/plugins/hanlp/plugin-security.policy
7.修改/home/elasticsearch-5.6.3/bin/elasticsearch.in.sh,修改ES_CLASSPATH
ES_CLASSPATH="$ES_HOME/lib/*:$ES_HOME/plugins/hanlp/"
8.分词效果验证
进入Dev Tools,输入以下文本:
GET /_analyze?analyzer=hanlp&pretty=true
{
"text":"记者从四川省农业农村厅了解到"
}
输出:
{
"tokens": [
{
"token": "记者",
"start_offset": 0,
"end_offset": 2,
"type": "nnt",
"position": 0
},
{
"token": "四川省",
"start_offset": 3,
"end_offset": 6,
"type": "ns",
"position": 1
},
{
"token": "农业",
"start_offset": 6,
"end_offset": 8,
"type": "n",
"position": 2
},
{
"token": "农村",
"start_offset": 8,
"end_offset": 10,
"type": "n",
"position": 3
},
{
"token": "厅",
"start_offset": 10,
"end_offset": 11,
"type": "n",
"position": 4
},
{
"token": "了解到",
"start_offset": 11,
"end_offset": 14,
"type": "v",
"position": 5
}
]
}
以上就是es和hanlp插件的安装使用,后面还会介绍如何建立分词索引和从mysql向es导入文本。
五.参考文档
https://www.jianshu.com/p/52c42cdab997
https://blog.csdn.net/cb2474600377/article/details/78963247