NLPIR/ICTCLA2018 ACADEMIC SEMINAR 8th ISSUE COMPLETED

NLPIR/ICTCLA2018 ACADEMIC SEMINAR 8th ISSUE COMPLETED自然语言处理与信息检索共享平台-yu7e
l#@){7eH

 自然语言处理与信息检索共享平台&?3nm-Ce{3t
fP

        Last week, Shen Li gave a presentation about the paper, Design Challenges and Misconceptions in Neural Sequence Labeling and shared some opinion on it.自然语言处理与信息检索共享平台p ~
` ^fmQJ/M3P+I#L

 自然语言处理与信息检索共享平台
k.ur)R:o4{i

        After the presentation, several questions were asked. The Q&As are listed as following:

*TqDbB,s\3d0

        Q: What is oov, oobv and ooev?自然语言处理与信息检索共享平台,sh7?
Y*X,gS?H%hz

        A: oov is Out-of-vocabulary, oobv is out-of-both-vocabulary, ootv is out-of-training-vocabulary, ooev is out-of-embedding-vocabulary.自然语言处理与信息检索共享平台Z5a6I(o L~1f[

        Q: How CNN and LSTM work?


E,j\z1Zs#^I0Kj p0

        A: The authors constructed a neural sequence labeling framework with three layers , a character sequence representation layer, a word sequence representation layer and an inference layer. LSTM and CNN are used on the character and word level, to caputure sequence representation information without hand-engineered features. More detailed information can be seen in figure 2 and 3自然语言处理与信息检索共享平台:]3D7jBvY%Y(~(A

 


aJ|4pTh
j5W.E0

        The powerpoint of the presentation can be download.自然语言处理与信息检索共享平台7E@ L’M1bu!h

 自然语言处理与信息检索共享平台oN/{+J4`b8vD
r

You May Also Like

About the Author: nlpir

发表回复