How to Extract Words Following a Key Word

Question

I'm currently trying to extract 4 words after "our", but keep getting words after "hour" and "your" as well.

ie) "my family will send an email in 2 hours when we arrive at." (text in the column)

What I want: nan (since there is no "our")

What I get: when we arrive at (because hour as "our" in it)

I tried the following code and still have no luck.

our = 'our\W+(?P<after>(?:\w+\W+){,4})' 
Reviews_C['Review_for_Fam'] =Reviews_C.ReviewText2.str.extract(our, expand=True)

Can you please help?

Thank you!

Answer 1

Im suprised to see regex used for this due to it causing unneeded complexity sometimes. Could something like this work?

def extract_next_words(sentence):
    # split the sentence into words
    words = sentence.split()
    
    # find the index of "our"
    index = words.index("our")

    # extract the next 4 words
    next_words = words[index+1:index+5]

    # join the words into a string
    return " ".join(next_words)

Answer 2

You need to make sure "our" is with space boundaries, like this:

our = '(^|\s+)our(\s+)?\W+(?P<after>(?:\w+\W+){,4})'

specifically (^|\s+)our(\s+)? is where you need to play, the example only handles spaces and start of sentence, but you might need to extend this to have quotes or other special characters.

Answer 3

Here is the generic code for finding the n number of words after a specific 'x' word in the string. It also accounts for multiple occurrences of 'x' as well as for non-occurrence.

def find_n_word_after_x(in_str, x, n):
    in_str_wrds = in_str.strip().split()
    x = x.strip()
    if x in in_str_wrds:
        out_lst = []
        for i, i_val in enumerate(in_str_wrds):
            if i_val == x:
                if i+n < len(in_str_wrds):
                    out_str = in_str_wrds[i+1:i+1+n]
                    out_lst.append(" ".join(out_str))
        return out_lst
    else:
        return []
str1 = "our w1 w2 w3 w4 w5 w6"
str2 = "our w1 w2 our w3 w4 w5 w6"
str3 = "w1 w2 w3 w4 our w5 w6"
str4 = "w1"

print(find_n_word_after_x(str1, 'our', 4))
print(find_n_word_after_x(str2, 'our', 4))
print(find_n_word_after_x(str3, 'our', 4))
print(find_n_word_after_x(str4, 'our', 4))

Generated Output:

['w1 w2 w3 w4']
['w1 w2 our w3', 'w3 w4 w5 w6']
[]
[]