如何计算文本中每对句子之间的距离

Question

i am computing the levenshtein distance between the sentences, and now i have a text with several sentences. I don't know how to write the for loop to generate the distance between each pair of the sentence.

sent = ['motrin 400-600 mg every 8 hour as need for pai . ', 'the depression : continue escitalopram ; assess need to change medication as an outpatient . ', 'Blood cltures from 11-30 grow KLEBSIELLA PNEUMONIAE and 12-01 grow KLEBSIELLA PNEUMONIAE and PROTEUS MIRABILIS both sensitive to the Meropenam which she have already be receive . ']

def similarity(sent):
    feature_sim = []
    for a,b in sent:
            feature_sim[a,b] = pylev.levenshtein(a,b) 
        print (feature_sim)

Answer 1

Use a pair of nested for-loops.

Simplest version:

for a in sent:
    for b in sent:
        ...

Skip identical pairs (Levenshtein distance would trivially be 0):

for a in sent:
    for b in sent:
        if a != b:
           ...

Avoid processing duplicate pairs ( a, b is the same as b, a ):

for i in range(0, len(sent)):
    for j in range(i+1, len(sent)):
        # a = sent[i], b = sent[j]
        ...

Problem: feature_sim is a list , which can only be indexed by integers, not strings or any other types.

Use a dictionary instead:

feature_sim = {}
for i in range(0, len(sent)):
    for j in range(i+1, len(sent)):
        feature_sim[(sent[i], sent[j])] = pylev.levenshtein(sent[i], sent[j])