在使用senticNet的时候,输入单个单词disgusting有反应,能判断出是一个消极词,但是在输入一整句话的时候,就没办法检测到disgusting,后来debug的时候方向,遍历句子中的单词的时候是"disgusting."
估计就是这个原因使得词典判断句子情感极性不准确。
解决方式:
line ='This place is disgusting, absolutely horrible, it is my second stay here, there is a strange stain on the other side of the bed I stay on, and my friends and I are rendered air conditionless, but it is 55 for downtown Chicago, what am I suppose to do?'
# print(line)
line=re.sub(r'(?<=[.,?])(?=[^\s])', r' ', line[::-1])#在标点符号句号和逗号后面加一个空格
line=line[::-1]#This is a text .This is another text ,it has no space after the comma .
wordlist = line.split(' ')
print(line)
print(wordlist)
运行结果:
D:\appsbag\Anpython\envs\pytorch\python.exe D:/Python/readXlsx/sentic.py
This place is disgusting , absolutely horrible , it is my second stay here , there is a strange stain on the other side of the bed I stay on , and my friends and I are rendered air conditionless , but it is 55 for downtown Chicago , what am I suppose to do ?
['This', 'place', 'is', 'disgusting', ',', 'absolutely', 'horrible', ',', 'it', 'is', 'my', 'second', 'stay', 'here', ',', 'there', 'is', 'a', 'strange', 'stain', 'on', 'the', 'other', 'side', 'of', 'the', 'bed', 'I', 'stay', 'on', ',', 'and', 'my', 'friends', 'and', 'I', 'are', 'rendered', 'air', 'conditionless', ',', 'but', 'it', 'is', '55', 'for', 'downtown', 'Chicago', ',', 'what', 'am', 'I', 'suppose', 'to', 'do', '?']
Process finished with exit code 0
参考资料:
Python Regex在点或逗号后添加空格