學習筆記之天然語言處理（Natural Language Processing）

天然語言處理 - 維基百科，自由的百科全書

https://zh.wikipedia.org/wiki/%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86

---

BEST PRACTICE

語音識別技術簡史

• https://mp.weixin.qq.com/s/wnPAnOaB0ydahZP-Da4Plw

NLP領域預訓練模型的現狀及分析

• https://mp.weixin.qq.com/s/vFsJE81Rs8C1zKoNv3K-bA

NLP學習思惟導圖，很是的全面和清晰

• https://mp.weixin.qq.com/s/e7Lp5Suzp1Z9P-ESExGJkw
• https://github.com/graykode/nlp-roadmap

練手紮實基本功必備：非結構文本特徵提取方法

• https://mp.weixin.qq.com/s/ghfh1x_lsEcoA8PFPXE46w

乾貨|一文讀懂詞性標註

• https://mp.weixin.qq.com/s/voXzXiJ-xQdvn6cc1ifz6A

NLP三種中文分詞工具的對比使用

• https://mp.weixin.qq.com/s?__biz=MzAxMjUyNDQ5OA==&mid=2653560767&idx=1&sn=e949a2cb82cbc357b5c5e2ed04a73f23&chksm=806e3102b719b814e9fd6498805e48db5a3a688aa596f67b6116efe165c76cac819452a894c5&mpshare=1&scene=24&srcid=&sharer_sharetime=1567993236268&sharer_shareid=5ed4a849fa42d9599a974fa8eb45e8fa&key=6d90834972a32f5a3032ba6357f32311f48632f518b1df7a5075f92c15110a43e1927caa99bb9b234b7b57d518e01ae28efe378f3835601a663a919a79775fd02005ef5189b019430c7b079691204ec9&ascene=14&uin=MTMzMzc3MjY4MQ%3D%3D&devicetype=Windows+10&version=62060833&lang=en&pass_ticket=tT3maEfznKd3xtVT4L8%2Bl%2B2KKdhrJZ3ERaWEoIpqIMB2I2ssKo%2BTfx0v80L7rMTL

簡單有效的多標準中文分詞

• https://mp.weixin.qq.com/s/S2xfakhh1EpqW_s8AFTNuA
• http://www.hankcs.com/nlp/segment/multi-criteria-cws.html

乾貨|一文讀懂依存句法分析

• https://mp.weixin.qq.com/s/0-g2AVUWe1fYTl8ZkCJNPQ
• https://github.com/yuquanle/StudyForNLP/blob/master/NLPbasic/Dependency.ipynb

框架|最新最早進的天然語言處理庫介紹，涵蓋BERT，GPT-2等

• https://mp.weixin.qq.com/s/Nqt7ALFOlg78WJ2sgNMstg

解密 BERT

• https://mp.weixin.qq.com/s/SPfa17p3QetZXCC01DwmQA
• https://www.analyticsvidhya.com/

BERT 瘦身之路：Distillation，Quantization，Pruning

• https://mp.weixin.qq.com/s/Soiq0zcH7tR92mmzXK5_eg

BERT 的演進和應用

• https://mp.weixin.qq.com/s/r0Dew1W2goLKdnYQOwdcJw
• https://zhuanlan.zhihu.com/p/72805778

將BERT思想用於語音識別！滴滴提出新無監督預訓練框架，中文識別性能提高10%

• https://mp.weixin.qq.com/s/bsRVFjhWg0s6GtyllJZ4-w
• https://arxiv.org/pdf/1910.09932.pdf

XLNet預訓練模型，看這篇就夠了！

• https://mp.weixin.qq.com/s/_8hcLR9y1c8RmjPb79Rw_g
• https://github.com/NLP-LOVE/ML-NLP
• https://github.com/ymcui/Chinese-PreTrained-XLNet

T5 模型：NLP Text-to-Text 預訓練模型超大規模探索

• https://mp.weixin.qq.com/s/0EDN-ILeL_diZ1G11ikwjw

知識蒸餾（Knowledge Distillation）簡述

• https://mp.weixin.qq.com/s/TM9_1f0hHtMZnu2dGe0s_A
• https://mp.weixin.qq.com/s/pXoENwz4Z-eok9y3P9rQvg
• https://zhuanlan.zhihu.com/p/82129871
• https://zhuanlan.zhihu.com/p/81467832

經驗|初入NLP領域的一些小建議

• https://mp.weixin.qq.com/s/at7_OOf8BSWlqO5eawzOQw
• https://zhuanlan.zhihu.com/p/59184256

---

APPLICATION

NLP 學習：從文本中提取時間

• https://mp.weixin.qq.com/s/F6nriPskzZ8g-L9ntRwvkQ