利用贝叶斯分析唐诗的作者是李白还是杜甫

本文链接：https://blog.youkuaiyun.com/vincent_duan/article/details/117992548

本文介绍了一个使用朴素贝叶斯分类器进行古诗词作者识别的小型项目。该项目通过jieba分词对李白和杜甫的诗歌进行预处理，并利用nltk库进行训练。用户可以输入诗句，系统将判断其更可能出自哪位诗人。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

import jieba
from nltk.classify import NaiveBayesClassifier

# 需要提前把李白的诗收集一下，放在libai.txt文本中。
text1 = open(r"libai.txt", "rb").read()
list1 = jieba.cut(text1)
result1 = " ".join(list1)
# 需要提前把杜甫的诗收集一下，放在dufu.txt文本中。
text2 = open(r"dufu.txt", "rb").read()
list2 = jieba.cut(text2)
result2 = " ".join(list2)

# 数据准备
libai = result1
dufu = result2


# 特征提取
def word_feats(words):
    return dict([(word, True) for word in words])


libai_features = [(word_feats(lb), 'lb') for lb in libai]
dufu_features = [(word_feats(df), 'df') for df in dufu]
train_set = libai_features + dufu_features
# 训练决策
classifier = NaiveBayesClassifier.train(train_set)

# 分析测试
sentence = input("请输入一句你喜欢的诗：")
print("\n")
seg_list = jieba.cut(sentence)
result1 = " ".join(seg_list)
words = result1.split(" ")

# 统计结果


lb = 0
df = 0
for word in words:
    classResult = classifier.classify(word_feats(word))
    if classResult == 'lb':
        lb = lb + 1
    if classResult == 'df':
        df = df + 1

# 呈现比例
x = float(str(float(lb) / len(words)))
y = float(str(float(df) / len(words)))
print('李白的可能性：%.2f%%' % (x * 100))
print('杜甫的可能性：%.2f%%' % (y * 100))