NLTK：语料库级别的 BLEU 与句子级别的 BLEU 分数

Question

我已经在 python 中导入了 nltk 来计算 Ubuntu 上的 BLEU 分数。我了解句子级 BLEU 评分的工作原理，但我不了解语料库级 BLEU 评分的工作原理。

以下是我的语料库级 BLEU 分数代码：

import nltk

hypothesis = ['This', 'is', 'cat'] 
reference = ['This', 'is', 'a', 'cat']
BLEUscore = nltk.translate.bleu_score.corpus_bleu([reference], [hypothesis], weights = [1])
print(BLEUscore)

由于某种原因，上述代码的 bleu 分数为 0。我预计语料库级别的 BLEU 分数至少为 0.5。

这是我的句子级 BLEU 分数代码

import nltk

hypothesis = ['This', 'is', 'cat'] 
reference = ['This', 'is', 'a', 'cat']
BLEUscore = nltk.translate.bleu_score.sentence_bleu([reference], hypothesis, weights = [1])
print(BLEUscore)

这里的句子级 BLEU 分数是我预期的 0.71，考虑到简洁性惩罚和缺少单词“a”。但是，我不明白语料库级别的 BLEU 分数是如何工作的。

任何帮助将不胜感激。

Answer 1

我们来看看：

>>> help(nltk.translate.bleu_score.corpus_bleu)
Help on function corpus_bleu in module nltk.translate.bleu_score:

corpus_bleu(list_of_references, hypotheses, weights=(0.25, 0.25, 0.25, 0.25), smoothing_function=None)
    Calculate a single corpus-level BLEU score (aka. system-level BLEU) for all 
    the hypotheses and their respective references.  

    Instead of averaging the sentence level BLEU scores (i.e. marco-average 
    precision), the original BLEU metric (Papineni et al. 2002) accounts for 
    the micro-average precision (i.e. summing the numerators and denominators
    for each hypothesis-reference(s) pairs before the division).
    ...

你比我更了解算法的描述，所以我不会试图向你“解释”它。如果文档字符串不够清楚，请查看the source 本身。或者在本地找到：

>>> nltk.translate.bleu_score.__file__
'.../lib/python3.4/site-packages/nltk/translate/bleu_score.py'

Answer 2

TL;DR：

>>> import nltk
>>> hypothesis = ['This', 'is', 'cat'] 
>>> reference = ['This', 'is', 'a', 'cat']
>>> references = [reference] # list of references for 1 sentence.
>>> list_of_references = [references] # list of references for all sentences in corpus.
>>> list_of_hypotheses = [hypothesis] # list of hypotheses that corresponds to list of references.
>>> nltk.translate.bleu_score.corpus_bleu(list_of_references, list_of_hypotheses)
0.6025286104785453
>>> nltk.translate.bleu_score.sentence_bleu(references, hypothesis)
0.6025286104785453

（注意：您必须在 develop 分支上拉取最新版本的 NLTK 才能获得稳定版本的 BLEU 分数实现）

---

长期：

实际上，如果整个语料库中只有一个引用和一个假设，corpus_bleu() 和 sentence_bleu() 都应该返回与上例所示相同的值。

在代码中，我们看到sentence_bleu is actually a duck-type of corpus_bleu:

def sentence_bleu(references, hypothesis, weights=(0.25, 0.25, 0.25, 0.25),
                  smoothing_function=None):
    return corpus_bleu([references], [hypothesis], weights, smoothing_function)

如果我们查看sentence_bleu 的参数：

 def sentence_bleu(references, hypothesis, weights=(0.25, 0.25, 0.25, 0.25),
                      smoothing_function=None):
    """"
    :param references: reference sentences
    :type references: list(list(str))
    :param hypothesis: a hypothesis sentence
    :type hypothesis: list(str)
    :param weights: weights for unigrams, bigrams, trigrams and so on
    :type weights: list(float)
    :return: The sentence-level BLEU score.
    :rtype: float
    """

sentence_bleu 引用的输入是 list(list(str))。

所以如果你有一个句子字符串，例如"This is a cat"，您必须对其进行标记以获得字符串列表["This", "is", "a", "cat"]，并且由于它允许多个引用，因此它必须是字符串列表的列表，例如如果您有第二个参考“这是一只猫”，您对sentence_bleu() 的输入将是：

references = [ ["This", "is", "a", "cat"], ["This", "is", "a", "feline"] ]
hypothesis = ["This", "is", "cat"]
sentence_bleu(references, hypothesis)

说到corpus_bleu()list_of_references参数，基本就是a list of whatever the sentence_bleu() takes as references：

def corpus_bleu(list_of_references, hypotheses, weights=(0.25, 0.25, 0.25, 0.25),
                smoothing_function=None):
    """
    :param references: a corpus of lists of reference sentences, w.r.t. hypotheses
    :type references: list(list(list(str)))
    :param hypotheses: a list of hypothesis sentences
    :type hypotheses: list(list(str))
    :param weights: weights for unigrams, bigrams, trigrams and so on
    :type weights: list(float)
    :return: The corpus-level BLEU score.
    :rtype: float
    """

除了查看nltk/translate/bleu_score.py 中的doctest，您还可以查看nltk/test/unit/translate/test_bleu_score.py 中的单元测试，了解如何使用bleu_score.py 中的每个组件。

顺便说一下，由于sentence_bleu在(nltk.translate.__init__.py](https://github.com/nltk/nltk/blob/develop/nltk/translate/init.py#L21)中被导入为bleu，所以使用

from nltk.translate import bleu 

将等同于：

from nltk.translate.bleu_score import sentence_bleu

在代码中：

>>> from nltk.translate import bleu
>>> from nltk.translate.bleu_score import sentence_bleu
>>> from nltk.translate.bleu_score import corpus_bleu
>>> bleu == sentence_bleu
True
>>> bleu == corpus_bleu
False