当前位置： 首页 > wzjs >正文

自己做的网站怎么弄成app湛江seo网站管理

wzjs 2025/9/19 22:31:56
自己做的网站怎么弄成app,湛江seo网站管理,it行业怎么入门,将夜影院在线观看免费完整版安装拼音分词器 选择es版本对应的pinyin分词器版本 下载后解压，放到es的插件目录下 重启es 自定义分词器 拼音分词器——可选配置 1. 首字母处理配置 keep_first_letter (默认: true) 解释：是否提取每个汉字的首字母组合，用于支持首字母…

安装拼音分词器

选择es版本对应的pinyin分词器版本

下载后解压，放到es的插件目录下

重启es

自定义分词器

拼音分词器——可选配置

1. 首字母处理配置

keep_first_letter (默认: true)

解释：是否提取每个汉字的首字母组合，用于支持首字母缩写搜索
开启时：刘德华 → [ldh]
关闭时：刘德华 → []（不生成首字母）
应用场景：适用于"ldh"搜索"刘德华"的需求

keep_separate_first_letter (默认: false)

解释：是否将每个汉字的首字母分开存储
开启时：刘德华 → [l,d,h]
关闭时：刘德华 → [ldh]
注意：开启会增加索引体积，但能支持更灵活的搜索（如"l d h"）

limit_first_letter_length (默认: 16)

解释：限制首字母结果的最大长度
示例：
中华人民共和国 → 默认输出[zhrmghg]（7字符）
设置为3时 → [zhr]
用途：控制长文本的首字母结果长度

2. 完整拼音处理

keep_full_pinyin (默认: true)

解释：是否保留每个汉字的完整拼音
开启时：刘德华 → [liu,de,hua]
关闭时：刘德华 → []
必要性：支持拼音精确搜索的基础配置

keep_joined_full_pinyin (默认: false)

解释：是否将完整拼音连接成一个词
开启时：刘德华 → [liudehua]
关闭时：刘德华 → [liu,de,hua]
优劣：连接后减少索引词项，但会丢失单字搜索能力

3. 非中文处理配置

keep_none_chinese (默认: true)

解释：是否保留原始文本中的非中文字符
开启时：刘德华AT2016 → [liu,de,hua,AT2016]
关闭时：刘德华AT2016 → [liu,de,hua]
重要性：处理混合文本的关键参数

keep_none_chinese_together (默认: true)

解释：是否保持非中文连续字符的完整性
开启时：DJ音乐家 → [DJ,yin,yue,jia]
关闭时：DJ音乐家 → [D,J,yin,yue,jia]
影响：关闭后会显著增加索引词项数量

4. 高级处理配置

none_chinese_pinyin_tokenize (默认: true)

解释：是否将非中文按拼音规则拆分
开启时：liudehua2016 → [liu,de,hua,2,0,1,6]
关闭时：liudehua2016 → [liudehua2016]
特殊用途：处理拼音与数字混合的情况

remove_duplicated_term (默认: false)

解释：是否去除重复的词项
开启时：de的 → [de]
关闭时：de的 → [de,的]
权衡：节省30-50%索引空间，但影响高亮精度

keep_original (默认：false)

解释：是否保留原始的文本
开启时："北京" → ["北京", "beijing", "bj"]
关闭时："北京"→ ["beijing", "bj"]

5. 系统行为配置

ignore_pinyin_offset (默认: true)

解释：是否忽略拼音分词的位置偏移
开启时：允许重叠分词（节省资源）
关闭时：严格位置约束（保证高亮准确）
版本注意：Elasticsearch 6.0+必须关注此参数

自定义分词器的工作原理

elasticsearch中分词器（analyzer）的组成包含三部分:

  • character filter：在tokenizer之前对文本进行处理。例如删除字符、替换字符
  • tokenizer：将文本按照一定的规则切割成词条（term）。例如keyword,就是不分词;还有ik_smart
  • tokenizer filter：将tokenizer输出的词条做进一步处理。例如大小写转换、同义词处理、拼音处理等

案例

新建用于测试自定义分词器的索引库test

PUT /test
{"settings": {"analysis": {"analyzer": {"my_analyzer": {"tokenizer": "ik_max_word","filter": "py"  }},"filter": { "py": {"type": "pinyin","keep_full_pinyin": false, # 不保留每个汉字的完整拼音"keep_joined_full_pinyin": true, # 把完整的拼音连成一个长拼音"keep_original": true, # 保留原始的文本"limit_first_letter_length": 16, # 限制首字母的最大长度为16"remove_duplicated_term": true,  # 去除重复的选项"none_chinese_pinyin_tokenize": false  # 不将非中文按拼音规则拆分}}}},"mappings": {"properties": {"words": {"type": "text","analyzer": "my_analyzer","search_analyzer": "ik_max_word"}}}
}

创建倒排索引的时候使用 my_analyzer

查询的时候指定分词器为 ik_max_word

这样就不会出现查询"狮子"的时候，出现虱子有关的词条了

测试

POST /test/_analyze
{"text": ["了却君王天下事junwang天下事"],"analyzer": "my_analyzer"
}
{"tokens" : [{"token" : "了却","start_offset" : 0,"end_offset" : 2,"type" : "CN_WORD","position" : 0},{"token" : "leque","start_offset" : 0,"end_offset" : 2,"type" : "CN_WORD","position" : 0},{"token" : "lq","start_offset" : 0,"end_offset" : 2,"type" : "CN_WORD","position" : 0},{"token" : "君王","start_offset" : 2,"end_offset" : 4,"type" : "CN_WORD","position" : 1},{"token" : "junwang","start_offset" : 2,"end_offset" : 4,"type" : "CN_WORD","position" : 1},{"token" : "jw","start_offset" : 2,"end_offset" : 4,"type" : "CN_WORD","position" : 1},{"token" : "天下事","start_offset" : 4,"end_offset" : 7,"type" : "CN_WORD","position" : 2},{"token" : "tianxiashi","start_offset" : 4,"end_offset" : 7,"type" : "CN_WORD","position" : 2},{"token" : "txs","start_offset" : 4,"end_offset" : 7,"type" : "CN_WORD","position" : 2},{"token" : "天下","start_offset" : 4,"end_offset" : 6,"type" : "CN_WORD","position" : 3},{"token" : "tianxia","start_offset" : 4,"end_offset" : 6,"type" : "CN_WORD","position" : 3},{"token" : "tx","start_offset" : 4,"end_offset" : 6,"type" : "CN_WORD","position" : 3},{"token" : "事","start_offset" : 6,"end_offset" : 7,"type" : "CN_CHAR","position" : 4},{"token" : "shi","start_offset" : 6,"end_offset" : 7,"type" : "CN_CHAR","position" : 4},{"token" : "s","start_offset" : 6,"end_offset" : 7,"type" : "CN_CHAR","position" : 4},{"token" : "junwang","start_offset" : 7,"end_offset" : 14,"type" : "ENGLISH","position" : 5},{"token" : "天下事","start_offset" : 14,"end_offset" : 17,"type" : "CN_WORD","position" : 6},{"token" : "tianxiashi","start_offset" : 14,"end_offset" : 17,"type" : "CN_WORD","position" : 6},{"token" : "txs","start_offset" : 14,"end_offset" : 17,"type" : "CN_WORD","position" : 6},{"token" : "天下","start_offset" : 14,"end_offset" : 16,"type" : "CN_WORD","position" : 7},{"token" : "tianxia","start_offset" : 14,"end_offset" : 16,"type" : "CN_WORD","position" : 7},{"token" : "tx","start_offset" : 14,"end_offset" : 16,"type" : "CN_WORD","position" : 7},{"token" : "事","start_offset" : 16,"end_offset" : 17,"type" : "CN_CHAR","position" : 8},{"token" : "shi","start_offset" : 16,"end_offset" : 17,"type" : "CN_CHAR","position" : 8},{"token" : "s","start_offset" : 16,"end_offset" : 17,"type" : "CN_CHAR","position" : 8}]
}
PUT /test/_doc/1
{"words":"身上有虱子"
}PUT /test/_doc/2
{"words":"山里有狮子"
}

执行DSL

GET /test/_search
{"query": {"match": {"words": "虱子"}}
}

指定search_analyzer为ik_max_word前的结果

{"took" : 6,"timed_out" : false,"_shards" : {"total" : 1,"successful" : 1,"skipped" : 0,"failed" : 0},"hits" : {"total" : {"value" : 2,"relation" : "eq"},"max_score" : 0.33425623,"hits" : [{"_index" : "test","_type" : "_doc","_id" : "1","_score" : 0.33425623,"_source" : {"words" : "身上有虱子"}},{"_index" : "test","_type" : "_doc","_id" : "2","_score" : 0.3085442,"_source" : {"words" : "山里有狮子"}}]}
}

指定search_analyzer为ik_max_word后的结果

{"took" : 2,"timed_out" : false,"_shards" : {"total" : 1,"successful" : 1,"skipped" : 0,"failed" : 0},"hits" : {"total" : {"value" : 1,"relation" : "eq"},"max_score" : 0.9530773,"hits" : [{"_index" : "test","_type" : "_doc","_id" : "1","_score" : 0.9530773,"_source" : {"words" : "身上有虱子"}}]}
}

显然，第二个结果是才是我们所希望的。

自动补全

es提供了completion suggest 查询来实现自动补全的功能，这个查询会匹配用户输入开头的词条并返回。

参与补全查询的字段必须是completion类型的，字段里内容是参与补全的多个词条。

自动补全（DSL实现）

创建一个game索引库，里面仅有一个completion类型的字段——title

PUT /game
{"settings": {"analysis": {"analyzer": {"my_analyzer": {"tokenizer": "ik_max_word","filter": "py"  }},"filter": { "py": {"type": "pinyin","keep_full_pinyin": false,"keep_joined_full_pinyin": true,"keep_original": true,"limit_first_letter_length": 16,"remove_duplicated_term": true,"none_chinese_pinyin_tokenize": false}}}},"mappings": {"properties": {"title": {"type": "completion","analyzer": "my_analyzer","search_analyzer": "ik_max_word"}}}
}
POST /game/_bulk
{"index":{"_id":1}}
{"title":["原神","开放世界","角色扮演","动作冒险","多平台","米哈游"]}
{"index":{"_id":2}}
{"title":["王者荣耀","MOBA","5v5","竞技","手游"]}
{"index":{"_id":3}}
{"title":["绝地求生","大逃杀","FPS","射击","Steam"]}
{"index":{"_id":4}}
{"title":["英雄联盟","MOBA","PC","竞技","团队合作"]}
{"index":{"_id":5}}
{"title":["崩坏：星穹铁道","角色扮演","回合制","科幻","米哈游"]}

测试案例1

GET /game/_search
{"suggest": {"game_suggest": {"text": "mi","completion": {"field": "title","skip_duplicates":false,"size": 5}}}
}
{"took" : 0,"timed_out" : false,"_shards" : {"total" : 1,"successful" : 1,"skipped" : 0,"failed" : 0},"hits" : {"total" : {"value" : 0,"relation" : "eq"},"max_score" : null,"hits" : [ ]},"suggest" : {"game_suggest" : [{"text" : "mi","offset" : 0,"length" : 2,"options" : [{"text" : "米哈游","_index" : "game","_type" : "_doc","_id" : "1","_score" : 1.0,"_source" : {"title" : ["原神","开放世界","角色扮演","动作冒险","多平台","米哈游"]}},{"text" : "米哈游","_index" : "game","_type" : "_doc","_id" : "5","_score" : 1.0,"_source" : {"title" : ["崩坏：星穹铁道","角色扮演","回合制","科幻","米哈游"]}}]}]}
}

测试案例2

GET /game/_search
{"suggest": {"game_suggest": {"text": "ha","completion": {"field": "title","skip_duplicates":false,"size": 5}}}
}
{"took" : 0,"timed_out" : false,"_shards" : {"total" : 1,"successful" : 1,"skipped" : 0,"failed" : 0},"hits" : {"total" : {"value" : 0,"relation" : "eq"},"max_score" : null,"hits" : [ ]},"suggest" : {"game_suggest" : [{"text" : "ha","offset" : 0,"length" : 2,"options" : [ ]}]}
}

RestAPI实现自动补全

@Test
void testSuggest() throws Exception {
SearchRequest request = new SearchRequest("game");
request.source().suggest(new SuggestBuilder().addSuggestion("game_suggest", SuggestBuilders.completionSuggestion("title").prefix("mi").skipDuplicates(false).size(5)));SearchResponse response = client.search(request, RequestOptions.DEFAULT);
CompletionSuggestion completionSuggestion = response.getSuggest().getSuggestion("game_suggest");for (CompletionSuggestion.Entry entry : completionSuggestion.getEntries()) {for (CompletionSuggestion.Entry.Option option : entry) {// 获取补全文本String suggestedText = option.getText().string();// 获取关联文档的_source（如果有）Map<String, Object> source = option.getHit().getSourceAsMap();System.out.println("命中: " + suggestedText);System.out.println("关联文档: " + source);}
}
}
命中: 米哈游
关联文档: {title=[原神, 开放世界, 角色扮演, 动作冒险, 多平台, 米哈游]}
命中: 米哈游
关联文档: {title=[崩坏：星穹铁道, 角色扮演, 回合制, 科幻, 米哈游]}

http://www.dtcms.com/wzjs/796933.html

相关文章：

  • 讯代理网站网站建设需要上传数据库吗
  • 软件下载站网站源码免费js调用wordpress文章列表
  • 我想做个网站 详解怎么做作品设计思路范文
  • 远近互联网站建设网站 点击率
  • 防盗网站人做清洁软件开发者路线图
  • 学校网站建立WordPress grace7主题
  • cms仿站教程wordpress移动模块位置
  • 深圳网站建设公司的外文名是免费下载微信小程序
  • 济南网站建设认可搜点网络能网站右键禁止
  • 自助快速建站最简单的网页
  • 常州制作网站公司电子请柬网站开发
  • 中国移动官方网站官网合肥建设工程招聘信息网站
  • 什么网站需要经营性备案wordpress站点管理员
  • 设一个网站需要多少钱出售淘宝店铺的平台
  • 做网站有兼职吗d8 4.0 wordpress
  • 保健品网站怎么做的应用程序安装下载
  • 赣州网站建设方案工商执照注册号查询网
  • 书画工作室网站模板网站建设wordpress 多站点 固定链接
  • 北京南站属于哪个街道十堰网站建设报价
  • 怎么外贸网站推广ui培训内容
  • 定制建设网站官方网站下载6966
  • 自己做的网站如何管理做高铁在哪个网站买
  • 三水区建设局网站宝安中心做网站
  • 刘涛做的婴儿玩具网站重庆大型网络公司
  • 网站建设服务合同是否缴纳印花税长沙公司网络营销推广
  • 做软件代理去哪个网站电影网站做多大会有风险
  • net网站开发net网站开发电子商务网站建设步
  • 写作网站招聘邯郸网络推广服务平台
  • 网站建设赶集网dz 一步一步教你做网站
  • 网站建设苏州公司《网站开发与应用》试题