当用同义词检查两个字符串之间的相似性时，它几乎可以给我们带来超过肯定的准确性。例如

如果我们将两个字符串与word-net的同义词集进行比较，并使用wup_similarity方法，

word1= car
word2= horse

我们将几乎返回1.0。

我该如何克服这个问题？

我想收到“如果我比较汽车，我只想检查文字网络中的车辆，而不是动物中的车辆”

请给我任何建议或建议，谢谢。

for word1 in pre_process_text_MARKING_SCHEME:
    simi = []
    for word2 in pre_process_text_ANSWER_SHEET:
        sims = []
        syns1 = wordnet.synsets(word1)
        syns2 = wordnet.synsets(word2)

        for sense1,sense2 in product(syns1,syns2):

            d = wordnet.wup_similarity(sense1,sense2)
            if d != None:

                #print(d,"similarity between word ",sense1," and ",sense2)
                sims.append(d)

        if sims != []:
            max_sim = max(sims)
            #print("maximum similarity between",word1,word2," is ",max_sim )
            simi.append(max_sim)

    if simi != []:
        max_final = max(simi)
        final.append(max_final)

        #print(max_final,"max_final between ",word2)



##---------------Final Output---------------##

similarity_index = numpy.mean(final)


similarity_index = round(similarity_index,2)

print("Similarity index value : ",similarity_index)


if similarity_index > 0.8:
    print("Similar")
elif similarity_index >= 0.6:
    print("Somewhat Similar")
else:
    print("Not Similar")