I need to iterate over a list and add synonyms and hyponyms of the words back to the list. For example:
list_of_words = ["bird", "smart", "cool", "happy"]
list_of_words = list_of_words + list_of_words_synonyms + list_of_words_hypnonyms
I'm able to get the synonyms and hypnonyms for individual words, but need to iterate over a list of values.
s = wordnet.synset(word)[0]
needs to return a list with individual synonyms added to the original list.
Expected result is: list_of_words = ["bird", "smart", "cool", "happy", "hen", "cock"..other synonyms of bird, "clever", "intelligent", other synonyms of smart....and so on]
How can I get the synset function to iterate over the list_of_words and include these words in the list? I'm very new to text analysis. Any help is appreciated.
3 Answers
(create this new answer rather than update my existing one, as the question has updated quite a lot)
Finally understand what wordnet.sysets() returns by installing package "pattern" and doing a debugging. Here is the code that runs:
from pattern.en import wordnet
list_of_words = [u"bird", u"smart", u"cool", u"happy"]
list_of_words_synonyms = []
list_of_words_hypnonyms = []
for word in list_of_words:
sts = wordnet.synsets(word)
if len(sts):
st = sts[0]
list_of_words_synonyms.extend(st.synonyms)
list_of_words_hypnonyms.extend(hs.senses[0] for hs in st.hyponyms())
list_of_words = list_of_words + list_of_words_synonyms + list_of_words_hypnonyms
print(list_of_words)
Please notes:
- the duplication is not considered. If removing duplication is a requirement, then you may use sets.Set instead of list
- for each hypnonym, it has multiple senses. list_of_words_hypnonyms just includes the first one. If you want to include all of them, use below code to replace the corresponing line:
list_of_words_hypnonyms.extend(sense for hs in st.hyponyms() for sense in hs.senses) - for adding hypononyms to list_of_words_hypnonyms, generator expression is used
The result is:
[u'bird', u'smart', u'cool', u'happy', u'bird', u'smart', u'smarting', u'smartness', u'cool', u'dickeybird', u'cock', u'hen', u'nester', u'night bird', u'bird of passage', u'protoavis', u'archaeopteryx', u'Sinornis', u'Ibero-mesornis', u'archaeornis', u'ratite', u'carinate', u'passerine', u'nonpasserine bird', u'bird of prey', u'gallinaceous bird', u'parrot', u'cuculiform bird', u'coraciiform bird', u'apodiform bird', u'caprimulgiform bird', u'piciform bird', u'trogon', u'aquatic bird', u'twitterer']
Here is a quick implementation. Don't worry much about fakesynsets, which is just a mockup of wordnet.synsets. You can directly check the code following this function.
def fakesynsets(word):
from collections import namedtuple
sysnset = namedtuple('sysnset', ['synonyms', 'hyponyms'])
return [sysnset(synonyms = [word+'syn'+str(ii) for ii in range(1,3)], hyponyms = lambda : [word+'hyp'+str(ii) for ii in range(1,3)])]
list_of_words = ["bird", "smart", "cool", "happy"]
list_of_words_synonyms = []
list_of_words_hypnonyms = []
for word in list_of_words:
s = fakesynsets(word)[0]
list_of_words_synonyms.extend(s.synonyms)
list_of_words_hypnonyms.extend(s.hyponyms())
list_of_words = list_of_words + list_of_words_synonyms + list_of_words_hypnonyms
print(list_of_words)
EDIT: according to OP's comment. Output format has been changed.
Suppose you want an output like this:
result = [
["bird", "smart", "cool", "happy"],
[["Synonym 1 of bird...", ...], ["Synonym 1 of smart...", ...], ["Synonym 1 of cool...", ...], ["Synonym 1 of happy...", ...]],
...
]
New output format:
["bird", "smart", "cool", "happy", "synonym of bird", "hyponym of bird", "synonym of smart"... ]
You can iterate through the original word list as below:
from pattern.en import wordnet
list_of_words = ["bird", "smart", "cool", "happy"]
original_length = len(list_of_words)
for word in list_of_words:
s = wordnet.synsets(word)[0]
# append synonyms list to the result
list_of_words.append([s.synonyms])
# append hyponyms list to the result
list_of_words.append(s.hyponyms())
After the iteration, you can access the lists by:
for index in range(original_length):
print 'Displaying word %s' % list_of_words[index]
print 'Synonyms: %s' % str(list_of_words[index + original_length])
print 'Hyponyms: %s' % str(list_of_words[index + original_length + 1])
