Finding occurrences of words in text which are in a list words

Question

Possible Duplicate:
Check if multiple strings exist in another string

Say I have a list of allowed words/phrases:

'Stack'
'Overflow'
'Stack Overflow'
'Stack Exchange'
'Exchange'

and the following text to parse:

'Hello, and welcome to Stack Overflow. 
 Here are some words which should match: Stack, Exchange.'

I'd like to get the list of words which are found in the allowed list:

'Stack Overflow'
'Stack'
'Exchange'

What would be the best way to achieve the result?

The allowed list I'll be using could be at least a thousand words/phrases.

Seems like this answer from another question has some interesting pointers: http://stackoverflow.com/a/3261300/89391 — miku, Dec 30 '12 at 00:57
Do you know how what the largest number of words in a phrase will be? — Sam Mussmann, Dec 30 '12 at 02:30
Thanks everyone. I'll try out the possible solutions, though re-implementing grep in Python sounds a bit daunting. @SamMussmann if by phrase you mean the words/phrases in the list, then they'll be at max 4 words, with majority at 1-2 words. — john2x, Dec 30 '12 at 04:39

score 2 · Answer 1 · edited Dec 30 '12 at 02:07

2

Put the words in a list and after use

def intersect(x, y):
    return list(set(x) & set(y))
word_list_text=string.split(text)
words_found={}
words_found=intersect(word_list_text, words)

edited Dec 30 '12 at 02:07

greeness

15,956
5
50
80

answered Dec 30 '12 at 01:15

Mihai8

3,113
1
21
31

3

This solution will not find words containing space such as `Stack Overflow" – deinonychusaur Dec 30 '12 at 02:10

Pradyumna · Answer 2 · 2012-12-30T02:43:07.317

0

let words be the list of words you want to search and paragraph(the give paragraph of info) where you want to search the words for

for i in words:
    if i in paragraph:
        print i

this code is good for the paragraph and words in your question,but mind you this code will print stack if we have stackoverflow and no stack(individually) , this is both a advantage and a disadvantage ,depends on your purpose, if you want it for individual words use this

y=paragraph.split()
for i in words:
    if i in y:
        print i

edited Dec 30 '12 at 02:43

answered Dec 30 '12 at 01:16

Pradyumna

167
2
12

Shouldn't you use `in` instead of `__contains__`? `in` will call `__contains__` under the hood, and is more idiomatic. – Sam Mussmann Dec 30 '12 at 02:29

Hairr · Answer 3 · 2012-12-30T05:20:47.133

If you have the phrases as:

phrases = ['Stack','Overflow','Stack Overflow','Stack Exchange','Exchange']

then the text as:

text = """Hello, and welcome to Stack Overflow. 
Here are some words which should match:Stack, Exchange."""

The following then can be done:

found_words = [word for word in phrases if word in text]

This will then eliminate the other phrases if they are not in the text. Though this will also Overflow in addition to the ones specified.

Finding occurrences of words in text which are in a list words

3 Answers3

Linked