我有一个数据框，其中包含来自客户服务电话对话的音频转录列。我创建了一个包含单词和句子的列表
words = ["rain", "buy new house", "tornado"]
我需要做的是在数据框中创建一列，逐行检查文本列中的这些单词，如果出现，则用单词及其频率更新列。例如第一行文本
"I was going to buy new house last week but it was raining since then. Once the rain stops I'll go and buy new house"
该列应为
{"buy new house",2}, {"rain",2}
或者可以创建重复行并在下一行添加逗号部分。
由于我还很新，如何进行此操作。
这是处理核心逻辑的一种方法：
def count_phrases(string: str, phrases: str) -> dict:
"Find the number of occurances of phrases in a string."
return {phrase: string.count(phrase) for phrase in phrases}
string = "I was going to buy new house last week but it was raining since then. Once the rain stops I'll go and buy new house"
phrases = ["rain", "buy new house", "tornado"]
assert count_phrases(string, phrases) == {'rain': 2, 'buy new house': 2, 'tornado': 0}
然后该函数可以在 Pandas DataFrame 中使用.apply