Solr 高效自动联想

scuess

步骤1：

         首先创建三个字段：分别存放汉字，拼音和拼音首字母
<field name="hanzi" type="string" indexed="true" stored="false" multiValued="false"/>
<field name="pinyin" type="kw_pinyin" indexed="true" stored="false" multiValued="true"/>
<field name="py" type="kw_py" indexed="true" stored="false" multiValued="false" />


然后把汉子转化成拼音和拼音首字母
<copyField source="hanzi" dest="pinyin"/>
<copyField source="hanzi" dest="py"/>



步骤2：

  在schema.xml文件中添加如下配置。
<fieldType name="kw_pinyin" class="solr.TextField" positionIncrementGap="0">  
<analyzer type="index">
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_cn.txt" />  
<filter class="com.shentong.search.analyzers.PinyinTransformTokenFilterFactory" minTermLenght="2" />  
<filter class="com.shentong.search.analyzers.PinyinNGramTokenFilterFactory" minGram="1" maxGram="20" />  
</analyzer>  
<analyzer type="query">
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_cn.txt" />
</analyzer>  
</fieldType>  
<fieldType name="kw_py" class="solr.TextField" positionIncrementGap="0">
<analyzer type="index">
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_cn.txt" />
<filter class="com.shentong.search.analyzers.PinyinTransformTokenFilterFactory" isFirstChar="true" minTermLenght="2"/>
</analyzer>
<analyzer type="query">
<tokenizer class="solr.KeywordTokenizerFactory"/>
<filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_cn.txt" />
</analyzer>
</fieldType>


步骤3：

   通过q=hanzi:value* OR pinyin:value OR py:value&fl:hanzi

   如果想加权重还可以在q后面拼接字符串酒OK，比如 ：

         defType:edismax
         bf:abs(weight)^40

    让某个分类、品牌、点击量等优先显示的权重设置。