Compare two Dataframes of sentences and return a third one

问题

I want to compare two long Dataframe columns of sentences, and return a third dataframe that looks like this. a snapshot is shown below.

My first approach was long winded and only worked for single instances, but failed when i applied it to the dataframe. It can be found in a previous question.

The logic is for words in c1 and c2, new value =1, for word in only c1, value set to zero.

sentences = tra_df['Sent1']
context = tra_df['Sent2']

Sent1[0] = "I am completely happy with the plan you have laid out today"
Sent2[0] = 'the plan you have laid out today'
c3 = ['0', '0', '0', '0' , '0', '1', '1', '1', '1', '1', '1'] 

回答1:

According to my understanding of your question, here is the solution.

def get_common_words(c1, c2):
    res = [0]*len(c1.split())
    for idx, existing_word in enumerate(c1.split()):
        if existing_word in c2.split():
            res[idx] = 1
    return res

get_common_words(c1, c2)

If you want to make it work for a pandas dataframe

def get_common_words_df(row):
   c1 = row['Sent1']
   c2 = row['Sent2']
   return get_common_words(c1, c2)


df['sent3'] = df.apply(get_common_words_df, axis=1)

You can optimize it a lot

来源：https://stackoverflow.com/questions/61326907/compare-two-dataframes-of-sentences-and-return-a-third-one