91
These copy and pasted notebooks are available, which are plagiarism of my notebooks I published.
This even wrote in the title that it is a copy of mine.
・https://www.kaggle.com/code/pixyz0130/birdclef2022-kaerururu-s-infer-addtest
・https://www.kaggle.com/code/pixyz0130/birdclef2022-use-2nd-label-f1
・https://www.kaggle.com/code/lunapandachan/birdclef2022-use-2nd-label-f2
I spent a lot of ideas (some borrowed from past competitions) and time creating my original notebooks (train fold0, inference).
It's sad that these copy and pasted notebooks earn a lot of upvotes with less effort.
Also, The more notebooks which have no new ideas , the harder it is to find notebooks which have new ideas.
We should not make kaggle a place where people who cheat like this can get benefit.
No one will share their knowledge, if such people increase and these cheating are allowed.
If you agree, please report them.
Comments (16)
Sort by
Hotness
arrow_drop_down-52
Personally speaking I disagree with you.
I appreciate pixyz0130's works as well as your works.
Some of his or her notebooks were respectably much explained, while the copies of yours were not.
I guess that he or she simply couldn't understand your codes and so he or she just tested those.
Frankly speaking, your notebooks are relatively harder to read than such as 'starter' or 'EDA'.
Of course that's not you fault. That's simply because you've left your works not as a kind of those but just as your insightful attempt notes.
I declare that I admire you. However, to say the truth, I cannot fully understand your findings.
'Less efforts'?
No, in fact he or she did make 'more efforts' that you didn't make, efforts of introducing your excellent works to the Japanese kagglers' alliance. Though I know those of your copies sadly resulted to be just a confirming.
Please don't speak ill of others, you Competition Master.
The contributors that you look as a plagiarist may benefit you non-directly.
Just like Hiroyuki's video pick-uppers, who finally make Hiroyuki more popular.
個人的には賛同できません。
私は、あなたに感謝するのと同様に、あなたが挙げたpixyz0130にも感謝しています。
この方が出したノートブックは、あなたのノートブックのコピー以外では割とちゃんと解説されています。
思うに、単純にあなたのコードが理解できず、ただテストするにとどまったのではないでしょうか？
率直に言えばあなたのノートブックはEDAやスターターと比べて決して読みやすくありません。
それはあなたの落ち度ではなくて、単純に、そういう体では公開されておらず、「書き置き」という形で残したからです。
私はあなたの貢献に感謝していますが、正直あなたの試行をあまり理解できていません。
「努力していない」というのはどうかと思います。
実際には、あなたがしていなかった「日本語Kagglerへの紹介」という努力をしていると思いますよ。
くだんのノートに関しては残念ながら「追試」にしかなっていなかったかもしれませんが…
ともあれ、Competiton Masterの方があまり悪口を言わないでいただきたいです。
剽窃者に見える人も、直接的ではないにせよあなたの利益になる可能性がありますよ。
ひろゆき動画を切り貼りして稼いでいる人たちが詰まるところ彼の知名度を上げているように。
3
Agree with you, a gold medal notebook should be original.
5
As far as I am concerned, it should be reasonable to apply the practice of cross refence/citation when we borrow wisdom from others.
After all, the original kernel creators should take the most credits.
Maybe a little way too academic? Lol.
10
Thank you for sharing your opinion, I was thinking the same thing.
In addition, I think those notebooks are unuseful in that they are written in Japanese, and also nonsense in that they are using Japanese inside joke = Yukkuri Jikkyou; contents frequently used in Japanese Youtube channel.
For sure, Notebooks and Discussion written in our mother's tongue is easy to read, but they may interrupt the role of
information exchange. (So I'm writing discussion in my poor English like this comment.)
I think Notebook downvote should be introduced.
2
I agree that plagiarism is not acceptable. It is not conducive to a positive environment, is detrimental to the community, and should be discouraged. Thinking and coming up with unique work is hard work and having someone straight up copying your work and receiving all the benefits without any regard for recognition of the original source is pretty infuriating. That said, I believe our open source world offers wonderful opportunities to learn and improve which involves building on top of what others have done before us; however, best efforts should be put towards giving credit where credit is due. I think this spirit helps the community grow and fosters a better environment where we can and add to the development ourselves and others.
1
Speaking of which,
your notebook borrows a lot of code from https://github.com/Ajax0564/Birdcall-Sound-Event-Detection/ without citation
3
I support you. we should respect every original author!
-25
I disagree with calling Derivative Works under Apache License 2.0 "plagiarism" or "cheating".
- "This Notebook has been released under the Apache 2.0 open source license." https://www.kaggle.com/code/kaerunantoka/birdclef2022-ex005-f0-infer
- "This Notebook has been released under the Apache 2.0 open source license." https://www.kaggle.com/code/kaerunantoka/birdclef2022-use-2nd-label-f0
- "By so sharing, you are deemed to have licensed the shared code under an Open Source Initiative-approved license (see www.opensource.org)" https://www.kaggle.com/competitions/birdclef-2022/rules
kaerururuTopic Author • (34th in this Competition) • a day ago • Options •
Report Message
9
I agree with your opinion that these copies are not cheating "in rule" because my codes are released under the Apache 2.0 open source license.
Did you read these copied notebooks ?
No additional information in these copies. Just copy and pasted (and few japanese joke).
If these works are forgiven, no one will publish their codes at all.
Perhaps you recognize plagiarism under Community Guidelines as plagiarism under Apach 2.0.
The Community Guidelines prohibit plagiarism as follows. However, plagiarism here does not mean plagiarism under Apache 2.0.
Don’t try to manipulate the progression system.
Progression manipulation includes any behaviors that intend to game Kaggle’s reputation system. They harm the fairness and quality of the community and are forbidden on our platform. Off-topic self promotion, upvote collusion, plagiarism, and cheating are some (but not all) examples of progression manipulation.
Also, since the Community Guidelines do not explicitly state the criteria for plagiarism, I see nothing wrong with the step of reporting and asking Kaggle team to make a decision.
2
@kaerunantoka
I disagree with the offensive words, not your whole opinion.
I also think it would be better if ゆっくり実況 notebooks didn't exist.
@onodera @yukkyo
Thank you for letting me know and reminding the usage of "plagiarism" on Kaggle.
To my current understanding, "report plagiarism" means "report notebooks with the option Plagiarism/copied content that is not meaningfully different", and "plagiarism" on Kaggle doesn't necessarily mean "pretending that it is your own" https://dictionary.cambridge.org/dictionary/english/plagiarism .
0
I support your opinion. Gold-medal submission should be original, and creator should be evaluated with fair.
0
i agree with you..it's a fact