在Python中正确使用NTLK

Question

I am trying to use python with NTLK to get a number of acronyms for a number of words (for now 2). It seems that I can get it to work with the first word, but not the second. I'm guessing I still have lots to learn about NTLK. There is some simplified example code below. I am basically trying to get two lists of acronyms, 1 list for each word. All was well with the first for loop. After I tried the second word I get:

syn2 = wn.synsets(word)[0].lemmas[y]
IndexError: list index out of range

Hope someone can aid me in my understanding of why this is happening.

import nltk
from nltk.corpus import wordnet as wn
import string
from array import * 

syn1 = '' 
syn2 = '' 
mylist = []    
mylist2 = []    
mylist3 = []  


Web_Keywd = 'car loan'
wuser_words = Web_Keywd.split()   

for word in wuser_words:                           


     i=i+1
     #first
     if (i == 1) :
         synset1 = wn.synsets(word)    
         y = 0     
         for synset in synset1:
             syn1 = wn.synsets(word)[0].lemmas[y]
             syn1 = syn1.name
             mylist2.append(syn1)
             y=y+1
     if (i == 2) :
         y = 0     
         for synset2 in wn.synsets(word):
             syn2 = wn.synsets(word)[0].lemmas[y]
             syn2 = syn2.name
             mylist3.append(syn2)
             y=y+1  

Answer 1

I've perhaps misleaded you in my previous answer with the use of wn.synsets(word)[0].lemmas[y] . You need to explicitly loop over the lemmas, as you can't know how many there are in advance. Example use case:

Web_Keywd = 'car loan cheap'

results = {}
for word in Web_Keywd.split():
    for synset in wn.synsets(word):
        for lemma in synset.lemmas:
            results.setdefault(word, []).append(lemma.name)

results now looks as follows:

{'car': ['car', 'auto', 'automobile', 'machine'...],
'loan': ['loan', 'loanword', 'loan', 'lend', 'loan'...],
'cheap': ['cheap', 'inexpensive', 'brassy', 'cheap...]}

To get unique results for each word submitted, independently of the others :

.... # same as above
            results.setdefault(word, set()).add(lemma.name)

To get a list of unique words for all the words submitted:

Web_Keywd = 'car loan cheap'

words = set(Web_Keywd.split())
results = set(
    lemma.name
    for word in words
        for synset in wn.synsets(word)
            for lemma in synset.lemmas
)
# results -> {'loanword', 'tatty', 'automobile', 'cheap', 'chinchy',...