Solr评分关键词检测率

Question

我用了Solr 6.1

我正在设定分数，

但我在分数上有一些问题

我只是搜索GCS，qf集是:标题^100内容^70文本^50，

这三个字段类型都是text_general，

我得到的第一个结果是1050.8486，另一个是853.08655，

但是第一个内容在内容字段中是如此之短，而另一个在内容字段中是如此之多，

我只是不知道为什么第一个分数会很高

以下两个结果调试器内容：

1002.8741 =之和：\n 1002.8741 = max of:\n 1002.8741 =重量(标题:GCS in 1275年) []，结果：\n 1002.8741 =得分(doc=1275，freq=1.0 = termFreq=1.0\n)，乘积:n 100.0 = boost\n 8.513557 = tfNorm (docFreq=27，docCount=137000)按：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.75 =参数b\n 6.3423285 = avgFieldLength\n 4.0 = fieldLength\n 928.3479 =权重(内容:GCS in 1275) []计算，结果：\n 928.3479 =得分(doc=1275，freq=2.0 = termFreq=2.0\n)，乘积：\n 70.0 = boost\n 7.1785564 = idf(docFreq=104，docCount=137000)\n 1.8474623 = tfNorm，计算结果如下：\n 2.0 = termFreq=2.0\n 1.2 =参数k1\n 0.75 =参数b\n 176.37256 = avgFieldLength\n 16.0 =字段长度\n

811.1335 =之和：\n 811.1335 = max of:\n 127.21202 =权重(文本:GCS in 9400) []，结果为：\n 127.21202 =得分(doc=9400，freq=1.0 = termFreq=1.0\n)，乘积为:n 50.0 = boost\n 7.464645 = tfNorm (docFreq=78，docCount=137000)从：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.75 =参数b\n 44.69738 = avgFieldLength\n 256.0 = fieldLength\n 811.1335 =权重(标题:GCS in 9400) []，结果：\n 811.1335 =得分(doc=9400，freq=1.0 = termFreq=1.0\n)，N 100.0 = boost\n 8.513557 = idf(docFreq=27，docCount=137000)\n 0.9527551 = tfNorm，计算结果如下：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.75 =参数b\n 6.3423285 = avgFieldLength\n 7.111111 =fieldLength n 174.06395 =重量(含量:GCS in 9400) []，结果:n 174.06395 =得分(doc=9400，freq=7.0 = termFreq=7.0\n)，乘积：\n 70.0 = boost\n 7.1785564 =tfNorm(docFreq=104，docCount=137000) 0.34639663 = tfNorm，计算结果如下：\n 7.0 = termFreq=7.0\n 1.2 =参数k1\n 0.75 =参数b\n 176.37256 = avgFieldLength\n 7281.778 =字段长度

===========================================================================

我还有另一个问题，当我使用碎片omitNorms时，它不会工作吗？为什么？我发现短内容得分多到长内容吗？模式是相同的。

第一个是来自A集合的短内容，另一个是B集合和长内容：

1158.9161 =之和：\n 1158.9161 = max of:\n 1158.9161 =重量(标题:波音52601) []，结果为:n 1158.9161 =得分(doc=52601，freq=1.0 = termFreq=1.0\n)，乘积为: 100.0 = boost\n 11.589161 = tfNorm (docFreq=5，docCount=593568)计算结果如下：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.0 =参数b(字段省略的规范)\n 1085.6042 =重量(内容:波音52601) []，结果为：\n 1085.6042 =得分(doc=52601，freq=2.0 = termFreq=2.0\n)，乘积为：\n 70.0 = boost\n 11.279006 =docCount=593568(docFreq=7，docCount=593568)= tfNorm，计算结果如下：\n 2.0 = termFreq=2.0\n 1.2 =参数k1\n 0.0 =参数b(字段省略的范数)\n

1060.8777 =之和：\n 1060.8777 = max of:\n 433.1234 =重量(文本:波音39406) []，结果：\n 433.1234 =得分(doc=39406，freq=1.0 = termFreq=1.0\n)，乘积为：\n 50.0 = boost\n 8.662468 = tfNorm (docFreq=112，docCount=650450)计算结果如下：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.0 =参数b(外地省略的规范)\n 884.746 =重量(标题:波音39406) []，结果：\n 884.746 =得分(doc=39406，freq=1.0 = termFreq=1.0\n)，乘积：\n 100.0 = boost\n 8.84746 =docCount=650450(docFreq=93，docCount=650450)= tfNorm，计算结果如下：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.0 =参数b(字段省略的规范)\n 1060.8777 =重量(内容:波音39406) []，结果为：\n 1060.8777 =得分(doc=39406，freq=7.0 = termFreq=7.0\n)，乘积为：\n 70.0 = boost\n 8.069756 =docCount=650450(docFreq=203，docCount=650450)= tfNorm，计算值：\n 7.0 = termFreq=7.0\n 1.2 =参数k1\n 0.0 =参数b(字段省略的范数)

Answer 1

下划线相似度Solr6.1使用的是BM251。

这意味着字段值长度与平均字段长度相比非常重要。更具体的是，您使用的是dismax，并且只考虑最大值。因此，探索极大值：

第一文档Max:

1002.8741 =重量(标题:GCS in 1275年) []，结果：\n 1002.8741 =得分(doc=1275，freq=1.0 = termFreq=1.0\n)，乘积：\n 100.0 = boost\n 8.513557 =tfNorm(docFreq=27，docCount=137000)\n 1.177973 = tfNorm，计算结果如下：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.75 =参数b\n 6.3423285 =avgFieldLength 4.0 = fieldLength\n

第二个文档Max:

811.1335 =重量(标题:GCS in 9400) []，结果为：\n 811.1335 =得分(doc=9400，freq=1.0 = termFreq=1.0\n)，乘积：\n 100.0 = boost\n 8.513557 =tfNorm(docFreq=27，docCount=137000)\n 0.9527551 = tfNorm，计算结果如下：\n 1.0 = termFreq=1.0\n 1.2 =参数k1\n 0.75 =参数b\n 6.3423285 = avgFieldLength\n 7.111111 = fieldLength\n

因此，短的第一个文件标题就是赢家。您可以使用dismax/edismax来考虑其他因素，而不仅仅是maximum2。

问候

1

2