如何在Python中计算忽略位置的单词序列匹配百分比

给定以下句子序列：

sentence1 = 'Ram is eating'
sentence2 = 'is Ram  eating'
sentence3 = 'is Ram playing'
sentence4 = 'movie Ram watching is'

由于difflib.SequenceMatcher是按字符匹配，无法直接满足忽略单词位置的匹配需求，我们可以通过以下方法实现指定规则的匹配百分比计算：

计算规则 #

匹配百分比公式：

match% = (sentence1与sentence2中匹配的单词数量/sentence1的总单词数)*100

实现步骤 #

1. 将句子按空白字符拆分单词（自动处理连续空格）
2. 通过集合求交集，得到两个句子共有的单词数量
3. 代入公式计算百分比

代码实现 #

sentence1 = 'Ram is eating'
sentence2 = 'is Ram  eating'
sentence3 = 'is Ram playing'
sentence4 = 'movie Ram watching is'

def get_match_percent(s1, s2):
    words1 = s1.split()
    words2 = s2.split()
    # 计算共同单词数量
    common_words = set(words1) & set(words2)
    matched_count = len(common_words)
    total_words = len(words1)
    if total_words == 0:
        return 0.0
    return (matched_count / total_words) * 100

# 输出结果
total_s1 = len(sentence1.split())
match1_2 = len(set(sentence1.split()) & set(sentence2.split()))
print(f"sentence1与sentence2的匹配百分比 = {match1_2}/{total_s1} 即{get_match_percent(sentence1, sentence2):.2f}%")

match1_3 = len(set(sentence1.split()) & set(sentence3.split()))
print(f"sentence1与sentence3的匹配百分比 = {match1_3}/{total_s1} 即{get_match_percent(sentence1, sentence3):.2f}%")

match1_4 = len(set(sentence1.split()) & set(sentence4.split()))
print(f"sentence1与sentence4的匹配百分比 = {match1_4}/{total_s1} 即{get_match_percent(sentence1, sentence4):.2f}%")

运行结果 #

sentence1与sentence2的匹配百分比 = 3/3 即100.00%
sentence1与sentence3的匹配百分比 = 2/3 即66.67%
sentence1与sentence4的匹配百分比 = 2/3 即66.67%

内容的提问来源于stack exchange，提问作者k_p