| 2020-03-13 15:20 |
Lead: Time flies like an arrow of time flies like an arrow.
Time flies like an arrow.
Time flies like an arrow.
Boring study fun, do for natural language processing (NLP) is especially true of people. Richness of language often lead to ambiguous understanding of this aspect has brought great challenges to NLP research, but also to enjoy NLPer group which has brought many self-ridicule fun.
In everyday life, these NLPer sensitive to language and "special habit" often difficult road to ordinary people, microblogging is to provide a platform for these lovers together happily communicate with each other pleasure.
NLP famous scholar Professor Liu Qun (Huawei currently serves as chief scientist laboratory Noah's Ark semantic voice) is such a genuine love of the natural language of scholars, microblogging, he was an associate professor with Qinghua Liu Zhiyuan and natural language understanding too is # # difficult two large coffee.
emmm ......, from the micro-blog name can be seen on their favorite
Late last night, Professor Liu Qun micro-Bo made such a push (if he's concerned about the micro-Fair found that this is the norm),
Time flies when hi arrows or as vector?
How to use a word to prove that you learned NLP? This attracted a lot of onlookers NLPers and self-ridicule. Taken following parts:
Analysis: This sentence used to describe the syntax from semantics can exist independently; there is no problem on this sentence syntax, but the sense of completely unable to explain. (Qun)
Analysis: natural language processing parts of speech than most classic: king - man + woman = queen. From 2013 Tomas Mikolov published "Efficient estimation of word representations in vector space".
Analysis: ".. Time flies like an arrow Fruit flies like a banana" is a humorous saying, this is a classic case of ambiguous language. "Time flies like an arrow, fruit flies like a banana."
Analysis: China: We (the outbreak) fast over Europe: our (country / people) finished soon China: We (the epidemic situation) better Europe: our (number of patients) better (Feng Zhiwei)
Parsing: [CLS] is the first sign of BERT, and the end of every sentence must be [SEP].
Analysis: NLP courses on translation of a joke, how are you? why you? how old are you? How old are you? (Qun)
Analysis: The girl with a telescope to see the boy or the girl saw the boy with a telescope. Classic syntactic ambiguity.
Analysis: In the early days of machine translation research, people often cite an example to illustrate the enormity of the task of machine translation. In English "The spirit is willing but the flesh is weak.", Meaning "beyond their grasp." But at the time of a machine translation system will sentence the English translation to Russian, then translated back to English when they became "The Voltka is strong but the meat is rotten.", Meaning "vodka is strong, but the meat was rotten. " The literal sense, "spirit" (liquor) and "Voltka" (vodka) on the translation seems to be no problem, "flesh" and "meat" also has the meaning of the meat. So these two words in the sense of why nothing in it? The key problem is that in the process of translation, machine translation systems for English idiom does not understand, just translate literally, with the result miss is as good sweeps clean. (Liu Zhiyuan)
Parsing: emm, this is not introduced.
…… ……
……
Something like this, often occur in the micro-Bo Professor Liu Qun, he used this for fun.
In fact, one of the largest gathering of ridicule on the micro-Bo, is a hot topic # # too difficult to understand natural language:
There is more than some well-known natural language processing large coffee, for example, Liu Qun (this topic was first created in 2015), Ma Shaoping, Liu Zhiyuan, Qiuxi Peng, Chen Boxing, Wang Bin, Ma Jianqiang and so on.
Only those who truly love will always concerned about the ambiguity of natural language and enrich lives languages. We are here also cited recent days in part of the topic, the brain can burn a glimpse of this gathering,
"March 8, is the birthday of Lao Tzu"
"This is not to protect competitors, but to opponents of protection."
"China: fast busy working on our side."
"We pay to ingratiate themselves."
Return to the original question, so, you learned how to use a word NLP prove it?
