将形容词和副词转换为其名词形式
我正在为我的项目尝试使用 wordnet 进行词义消歧。作为该项目的一部分,我想将派生形容词或副词形式转换为其根名词形式。
比如
漂亮==>美丽 精彩==>想知道
我怎样才能实现这个目标?除了 wordnet 之外还有其他字典可以提供这种转换吗?
如果我能将形容词的确切含义映射到具有确切含义的名词形式,这对我来说将是一个额外的好处。这可能吗?
谢谢
I am experimenting with word sense disambiguation using wordnet for my project. As a part of the project, I would like to convert a derived adjective or an adverb form to it's root noun form.
For example
beautiful ==> beauty
wonderful ==> wonder
How can I achieve this? Is there any other dict other than wordnet that provides this kind of transformation?
It would be an added bonus for me if I can map the exact sense of the adjective word to its noun form with exact sense. Is that possible?
Thank you
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)
在 google 或 SO 中搜索“词干提取”和“词形还原”等术语,这些术语可能会帮助您获得所需的内容。
例如,转到 http://qaa.ath.cx/porter_js_demo.html 并输入“美丽”和“美丽”这两个词,你会发现它们都有相同的含义。
波特词干分析器本质上删除了英语中常见的后缀,因此绝不是确定的,但是一个很好的起点。
您可以认为映射到同一词干的单词在某种意义上是同义的。如果您也可以获取所有这些单词的词性标签,您将能够推断出美丽是形容词美丽的名词形式。
Search google or SO for terms like 'stemming' and 'lemmatization', these terms might help you get what you are looking for.
For example, go to http://qaa.ath.cx/porter_js_demo.html and enter the words 'beautiful' and 'beauty', and you will see they both stem to the same token.
Porter stemmer essentially removes common suffixes found in the english language, so is by no means definitive, but is a pretty good place to start.
You can consider words that map to the same stem to be in some sense synonymous. If you can procure part of speech tags for all these words as well, you will be able to infer that beauty is the noun form of the adjective beautiful.