Images, words and sounds are the three most common forms of human contact and understanding the external world, and words are the most easily preserved and exchanged forms. A large number of human civilizations are ultimately preserved in the form of words; a large number of information exchanges are carried out in the form of words. How can machines be able to communicate with people, understand human thoughts, and ultimately understand words and various ideas and intentions behind them like humans? Natural language processing, the so-called NLP, is an important supporting technology.

In traditional web offense and defense, everyone has formed an indissoluble bond with the http protocol. But in the security field, web offense and defense are only a small branch. Many obviously unrelated spam emails can be seen by people, but rules-based spam gateways are always unsatisfactory to process. More and more e-commerce, forums and even video websites can always see obvious personal attacks or illegal information, but the filtering mechanism based on rules is always bypassed. Humans can easily understand the twenty-four-way switch and know whether an apple is a fruit or a mobile phone, but how does the machine do it? The answer is NLP.
This may be the first NLP introductory book written with open source ideas, and the entire writing process is on my Github.
https://github.com/duoergun0729/nlp
The reason I want to write with the open source idea is mainly because NLP technology, especially NLP technology based on machine learning, is developing very fast. For example, the fasttext technology, which is widely used, was published in 2016 and entered the field of mass production in 2017. But many natural language processing books remain within the scope of university courses, and even word vectors are rarely involved. I can write and update paper books with relatively cumbersome cycles on Github, and can also be quickly modified if there is an errata. I have completed three of them at present, and I will continue to update the content later. You can subscribe to my Github, or follow my WeChat official account " Brother Dou takes you to learn safety "

© 2018~2020 Brother Dou.
This work is licensed under the Creative Commons Attribution-Noncommercial Use 4.0 International License. Any commercial conduct that uses the book without my permission is illegal.
Writing a book is not easy, especially writing using personal rest time. Thank you for your reward, 100 yuan is not too much, 1 yuan is not too little.

If you prefer the form of Knowledge Planet, you can join my Knowledge Planet, and NLP-related questions can also be asked in the planet.
