Week1-5Why is NLP hard?
来源:互联网 发布:淘宝上买鞋子是正品吗 编辑:程序博客网 时间:2024/05/21 10:20
Examples
Time flies like an arrow.
Interpretations
More Examples
Syntax vs. semantics
*Little a has Mary lamb
- Syntactically wrong, markd as *
?Colorless green idea sleeps furiously
- Syntactically right, but semantically wrong, marked as ?
- idea doesn’t sleep
- sleep furiously?
- colorless green?
- green idea?
Ambiguous words
- ball, board, plant
- meaning
- fly, rent, tape
- part of speech
- address, resent, entrance, number, unionized
- different meanings
Ambiguity
- Not in computer language(by design)!
- Lojban
- Noun-noun phrase: (XY)Z vs. X(YZ)
- science fiction writer
- state chess tournament
Types of ambiguity
- Morphological
- Phonetic
- Part of speech
- Syntactic
- PP attachment
- Sense
- Modality
- Subjectivity
- CC(coordinating conjunction) attachment
- Negation
- Referential
- Reflexive
- Ellipsis and parallelism
- Metonymy
Other sources of difficulties
- Non-standard, slang, novel words and usages
- Inconsistensies
- Typos and grammatical errors
- Parsing problems
- Complex sentences
- Counterfactural sentences
- Humor and sacarsm
- Implicature/inference/world knowledge
- Semantics and pragmatics
- Language is even hard for humans
Synonyms and paraphrases
0 0
- Week1-5Why is NLP hard?
- Top Five Reasons Why AJAX Is So Hard
- isArray: Why is it so bloody hard to get right?
- Functional Programming Is Hard, That's Why It's Good
- Mooc论文【Why is it so hard to learn programming?】
- 转载:7 Reasons Why Software Development Is So Hard
- Why is it so hard to make a Java program appear native?
- CHAPTER 5 Why are deep neural networks hard to train?
- Why Vector Clocks Are Hard
- coursera NLP学习笔记之week1最小编辑距离计算
- Why Ruby is Simple
- Why extends is evil
- Why is China angry?
- why is me?
- why gpDesc is NULL?
- WHY IS GREP
- Why LD_LIBRARY_PATH is bad
- Why cacheAsBitmap is bad!
- android oom分析
- Linux与BSD中TCP协议栈实现比较
- iOS中UISearchBar(搜索框)使用总结
- 第四周--猴子选大王
- C++ 的swap手法
- Week1-5Why is NLP hard?
- CCScroview用法
- Knockout应用开发指南 第七章:Mapping插件
- 第七周 项目2 建立链队算法库
- Spring源码分析
- 针对虚幻3引擎渲染底层的效率优化
- c# - string to byte[] and vice versa
- 16.1 目标跟踪的项目
- hdu 2504 又见GCD