1. Detecting Languages During Indexing
在索引的时候,solr可以使用langid UpdateRequestProcessor来识别语言,然后映射文本到特定语言的字段.solr支持这个功能的两个实现:
- Tika的语言解析功能:http://tika.apache.org/0.10/detection.html
- LangDetect语言解析:http://code.google.com/p/language-detection/
可以从 http://blog.mikemccandless.com/2011/10/accuracy-and-performance-of-googles.html中看到它们之间的对比.一般情况下,LangDetect支持更多的语言,具有更高的性能.
参考http://wiki.apache.org/solr/LanguageDetection获取更多的关于langid UpdateRequestProcessor信息.
1.1 Configuring Language Detection
可以在solrconfig.xml中配置langid UpdateRequestProcessor.两个实现具有相同的参数,最少,你需要指定语言识别的字段和字段的结果语言编码.
1.2 Configuring Tika Language Detection
这里是solrconfig.xml 中 Tika langid UpdateRequestProcessor的最小的配置.
<processor class="org.apache.solr.update.processor.TikaLanguageIdentifierUpdateProcessorFactory"> <lst name="defaults"> <str name="langid.fl">title,subject,text,keywords</str> <str name="langid.langField">language_s</str> </lst> </processor>
1.3 Configuring LangDetect Language Detection
这里是solrconfig.xml中最小的LangDetect langid配置.
<processor class="org.apache.solr.update.processor.LangDetectLanguageIdentifierUpdateProcessorFac tory"> <lst name="defaults"> <str name="langid.fl">title,subject,text,keywords</str> <str name="langid.langField">language_s</str> </lst> </processor>
1.4 langid Parameters
正如上面所提到的,两个langid UpdateRequestProcessor的实现具有相同的参数:
| 参数 | 类型 | 默认 | 必填 | 描述 |
| langid | Boolean | true | no | 开启/关闭语言解析 |
| langid.fl | string | none | yes | 逗号或者空格分隔的字段列表.用于语言探测解析. |
| langid.langField | string | none | yes | 对返回的语言代码指定字段 |
| langid.langsField | multivalued string | none | no | 对返回的语言代码指定字段.如果使用langid.map.individual,每一个解析的语言都被添加到这个字段. |
| langid.overwrite | Boolean | false | no | 指定langField和langsField字段的内容是否被重写.如果它们包含值的话. |
| langid.lcmap | string | none | false | 空格分隔的列表,指定冒号分隔的语言代码用于语言解析.举例,你可以能使用这个映射中文,日文,韩文到一个cjk字段,并且映射美国英语和英国英语到一个en代码.可以使用langid.lcmap=ja:cjk zh:cjk ko:cjk . This affects both the values put into the en_GB:en en_US:en.这使这两个值放入到langField和langsField字段中. |
| langid.threshold | float | 0.5 | no | Specifies a threshold value between 0 and 1 that the language identification score must reach before accepts it. With longer langid text fields, a high threshold such at 0.8 will give good results. For shorter text fields, you may need to lower the threshold for language identification, though you will be risking somewhat lower quality results. We recommend experimenting with your data to tune your results. |
| langid.whitelist | string | none | no | Specifies a list of allowed language identification codes. Use this in combination with to ensure that you only index langid.map documents into fields that are in your schema. |
| langid.map | Boolean | false | no | Enables field name mapping. If true, Solr will map field names for all fields listed in . langid.fl |
| langid.map.fl | string | none | no | A comma-separated list of fields for that is different langid.map than the fields specified in . langid.fl |
| langid.map.keepOrig | Boolean | false | no | If true, Solr will copy the field during the field name mapping process, leaving the original field in place. |
| langid..map.individual | Boolean | false | no | If true, Solr will detect and map languages for each field individually |
| langid.map.individual.fl | stromh | none | no | 逗号分割的字段列表,使用 langid.map.individual.不同于langid.fl中指定的字段. |
| langid.fallbackFields | string | none | no | If no language is detected that meets the score langid.threshold , or if the detected language is not on the , this langid.whitelist field specifies language codes to be used as fallback values. If no appropriate fallback languages are found, Solr will use the language code specified in . |
| langid.fallback | string | none | no | Specifies a language code to use if no language is detected or specified in . |
| langid.map.lcmap | string | determined by langid.lcmap |
no | A space-separated list specifying colon delimited language code mappings to use when mapping field names. For example, you might use this to make Chinese, Japanese, and Korean language fields use a common suffix, and map both American and British English *_cjk fields to a single by using *_en langid.map.lcmap=ja:cjk . zh:cjk ko:cjk en_GB:en en_US:en |
| langid.map.pattern | Java regular expression |
none | no | By default, fields are mapped as <field>_<language>. To change this pattern, you can specify a Java regular expression in this parameter. |
| langid.map.replace | Java replace | none | no | By default, fields are mapped as <field>_<language>. To change this pattern, you can specify a Java replace in this parameter. |
| langid.enforceSchema | Boolean | true | no | If false, the processor does not validate field names against langid your schema. This may be useful if you plan to rename or delete fields later in the UpdateChain |