确定一般陈述的积极性或消极性
我想知道一个推特是积极的还是消极的。 例如:
论文:乔纳斯兄弟吃木炭 “乔纳斯兄弟要吃木炭了”>>>正面
“乔纳斯兄弟与木炭无关”>>>否定
论点:梅尔马克被摧毁 “梅尔马克死了”>>>积极
“阿尔夫和他的朋友住在梅尔马克”>>>>>否定
我想知道是否有某种算法可以在通用的、非语言特定的基础上做到这一点。
最简单的方法是什么? 提前致谢。
I want to know if a twit is positive or negative.
For example:
thesis: jonas brothers eat charcoal
"Jonas Brothers are going to eat charcoal" >>> Positive
"Jonas Brothers have nothing to do with charcoal" >>> Negative
thesis: melmac is destroyed
"Melmac is dead" >>> positive
"Alf is living with his friends in melmac" >>> negative
I want to know if there's some kind of algorithm to do this, on a generic, non languaje specific basis.
What's the easiest approach?
Thanks in advance.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(1)
我认为我的好方法是保留否定语句的字典:不,不是,什么都没有,等等。在语句中找到这些关键字。您还必须通过计数系统来考虑双重否定。例如“我还没有给出答案”,是肯定的但包含否定词。
I think I good approach would be to keep a dictionary of negation statements : don't, isn't, nothing, etc. Find those keywords in a statement. You would also have to account for double negatives through a counting system. For instance " I haven't not given an answer", is positive but contains negative words.