该博客介绍了一个实现智能纠错的Python类ErrorRecovery，它利用N-gram和转换拼音的方法来纠正文本中的错误。通过组合相邻词汇生成N-gram，再将词汇转换为拼音，匹配字典中的正确词汇进行纠错。博客展示了如何处理句子中的错误，并提供了实际运行示例。
from basicInfo import BasicInfo
import jieba
from pypinyin import pinyin, lazy_pinyin
from dataServer import DataServer
class ErrorRecovery(object):
def __init__(self, dataSer):
self.basic = BasicInfo()
self.data = dataServer
""" ===============一定要带标点符号==========="""
def recoveryMain(self, sentence):
word_list, word_gram_list = self.gram2Main(sentence)
user_pin_list =[]
for item in word_gram_list:
pin_list=lazy_pinyin(item, errors='ignore')
user_pin_list.append(pin_list)
for k, v in dataServer.dict_gupiao_pin.items():
import jieba
from pypinyin import pinyin, lazy_pinyin
from dataServer import DataServer
class ErrorRecovery(object):
def __init__(self, dataSer):
self.basic = BasicInfo()
self.data = dataServer
""" ===============一定要带标点符号==========="""
def recoveryMain(self, sentence):
word_list, word_gram_list = self.gram2Main(sentence)
user_pin_list =[]
for item in word_gram_list:
pin_list=lazy_pinyin(item, errors='ignore')
user_pin_list.append(pin_list)
for k, v in dataServer.dict_gupiao_pin.items():
最低0.47元/天 解锁文章
899
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
