About NLP 100 Exercise
Change Log
2020-12-25
Released NLP 100 Exercise 2020 Rev 2.
- Completed Chinese translation
- All Chapters were translated by Runzhe Zhan and Youmi Ma.
- A bug fix in the Japanese version
- Revised a word usage in the problem 32 (#28)
2020-06-08
Released NLP 100 Exercise 2020 Rev 1.
- Completed English translation
- Chapters 5 to 10 were translated by Shun Kiyono, Benjamin Heinzerling, Ana Brassard, and Paul Reisert.
- Fixed a spelling mistake in the problem 07 (#12)
- Included a POS-tagged text in Chapter 4 (#13)
- Revised the description for the problem 38 (#14)
- Revised the description for the problem 08 (#24)
- Fixed an incorrect translation in the problem 18
- Other fixes for spelling and styling mistakes
- Korean translation (up to the problem 39)
- Chapters 1 to 4 were translated by Sangwhan Moon.
- Numerous bugfixes in the Japanese version
- Revised an expression in Chapter 8 (#1)
- Fixed an incorrect math rendering in the problem 49 (#2)
- Fixed an incorrect title in Chapter 6 (#2)
- Fixed the incorrect data in Chapter 4 (#6)
- Revised the description of the problem 51 (#7)
- Revised the descriptions in the problems 50 and 51 (#8)
- Revised the description of the problem 75 (#9)
- Fixed the incorrect link in Chapter 5 (#16)
- Revised the description of the problem 04 (#18)
- Changed the sample text in Chapter 5 to reduce errors in dependency parsers (#19)
- Revised the description of the problem 69 (#21)
- Revised the description of the problem 71 (#22)
We deeply appreciate those who involved in this release especially by reporting Issues and sending Pull Requests.
2020-04-06
Released NLP 100 Exercise 2020.
- Added problems about Deep Neural Networks.
- Multilingual support.
- Released English translations (Problems 00 to 39).
- Moved Chapter 6 (English text processing) of the old version into English version.
2015-03-12
Released NLP 100 Exercise 2015 (in Japanese only).
- Added problems about distributional representations.
- Released the data publicly associated with the problems.
- Revised the problems for more practical use of NLP.
2012-04-03
Released the first version (in Japanese only).
Data sources
- popular-names.txt: National data (converted to TSV format), distributed on the Web site, “Beyond the Top 1000 Names”, Social Security Administration (SSA).
- enwiki-country.json.gz: Articles (in JSON format) describing countries that were extracted from English Wikipedia dump as of April 5, 2020. This file is distributed under the term of Creative Commons Attribution-ShareAlike 3.0 Unported.
-
alice.zip: The novel “Alice’s Adventures in Wonderland” (written by Lewis Carroll) from Project Gutenberg. Please see
11.txt
for the Project Gutenberg License. - ai.en.zip: The article “Artificial intelligence” from English Wikipedia. This file is distributed under the term of Creative Commons Attribution-ShareAlike 3.0 Unported.
About NLP 100 Exercise
- NLP 100 Exercise was developed and maintained by Naoaki Okazaki (Tokyo Institute of Technology).
- Many thanks go to Shun Kiyono (Tohoku University) for suggestions and revisions, and Sho Yokoi (Tohoku University) and Ryo Takahashi (Tohoku University) for suggestions in developing NLP 100 Exercise 2020.
- NLP 100 Exercise was translated into English by Shun Kiyono, Benjamin Heinzerling, Ana Brassard, Paul Reisert, and Naoaki Okazaki.
- NLP 100 Exercise was translated into Korean by Sangwhan Moon.
- NLP 100 Exercise was initially developed by Naoaki Okazaki in 2011-2012 for the use in a study group for programming in Inui-Okazaki Laboratory (currently, Inui-Suzuki Laboratory).
- The format where new students for studying NLP solve 100 assignments was inspired by the practice in the laboratory of Prof Satoshi Sato, Nagoya University.
Contact
We welcome feedback/suggestions as issues and pull requests in the GitHub repository.
If you would like to contact us about a matter aside from feedback/suggestions, please send an email to nlp100
at nlp.c.titech.ac.jp
.