了解召回率和精度 [英] Understanding Recall and Precision

问题描述

我目前正在学习信息检索，而我更喜欢回忆和精确度的例子

I am currently learning Information retrieval and i am rather stuck with an example of recall and precision

搜索者使用搜索引擎来查找信息.第一个结果屏幕上有10个文档，第二个结果屏幕上有10个文档.

A searcher uses a search engine to look for information. There are 10 documents on the first screen of results and 10 on the second.

假设搜索引擎索引中有10个相关文档.

Assuming there is known to be 10 relevant documents in the search engines index.

Soo ...总共有20条搜索，其中10条是相关的.

Soo... there is 20 searches all together of which 10 are relevant.

有人可以帮助我理解这一点吗?

Can anyone help me make sense of this?

谢谢

推荐答案

召回率和精度可以衡量结果的质量.为了理解它们，我们首先定义结果的类型.您返回的列表中的文档可以是

Recall and precision measure the quality of your result. To understand them let's first define the types of results. A document in your returned list can either be

• 正确分类

• classified correctly

• 真实肯定(TP):与相关文档(真实)相关且确实已返回(真实)的文档
• 真实否定(TN):与文档无关(否定)且确实未返回(true)

分类错误

• 误报(FP):与文档无关但返回肯定的文件
• 假阴性(FN):相关但没有被否定的文件

那么精度是:

| TP | /(| TP | + | FP |)

|TP| / (|TP| + |FP|)

即确实相关的检索文档的比例

i.e. the fraction of retrieved documents which are indeed relevant

然后召回是:

| TP | /(| TP | + | FN |)

|TP| / (|TP| + |FN|)

即结果集中的相关文档所占的比例

i.e. the fraction of relevant documents which are in your result set

因此，在您的示例中，20个结果中有10个是相关的.这使您的精度为0.5.如果最多只剩下这10个相关文档，那么您的召回率为1.

So, in your example 10 out of 20 results are relevant. This gives you a precision of 0.5. If there are no more than these 10 relevant documents, you have got a recall of 1.

(在测量信息检索系统的性能时，仅考虑精度和召回率才有意义.您可以通过根本不返回任何结果(即，不返回任何虚假实例=>否FP)轻松地获得100％的精度.或通过返回每个实例来召回100％(即未丢失任何相关文档=>没有FN).

(When measuring the performance of an Information Retrieval system it only makes sense to consider both precision and recall. You can easily get a precision of 100% by returning no result at all (i.e. no spurious returned instance => no FP) or a recall of 100% by returning every instance (i.e. no relevant document was missed => no FN). )

这篇关于了解召回率和精度的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！