beer review results

Question

beer review results

yes1234man opened this issue 4 years ago · 5 comments

Hi,
thanks for sharing the codes. The main results of this paper are beer review results which is not shared, could you please assist me with sharing codes and data to reproduce these results. thank you for your collaboration.

Answer 1 · 2020-07-13T13:31:27.000Z

Please refer to the README for the instruction to obtain the data. The provided code should be sufficient for you to reproduce our results.

Answer 2 · 2020-07-13T13:57:59.000Z

Hi First of all, please do not close an error while this is not solved, this is unethical!! second, no this is not sufficient, before publication you argue you share your codes, afterwards you do not collaborate. How should we know if your results were really correct and not a fake one? In research community, everyone needs to share the code and data. Abh

…

On Mon, Jul 13, 2020 at 3:31 PM Shiyu Chang ***@***.***> wrote: Please refer to the README for the instruction to obtain the data. The provided code should be sufficient for you to reproduce our results. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AODM7IZATKWR7Z7ULI5YX2LR3MED7ANCNFSM4OYA3KTA> .

Answer 3 · 2020-07-13T14:48:16.000Z

We recently found that the beer review dataset has been removed by the original authors (https://snap.stanford.edu/data/web-BeerAdvocate.html). Due to this problem, we cannot publish the dataset at this point. You might consider contacting the original authors for a copy of the dataset. This is also unexpected to us. We could help to generate indices that match our partition once you have access to the original dataset.

Answer 4 · 2020-07-13T14:52:38.000Z

Hi Thanks for the explanations, but I cannot see why you could NOT share a copy of these datasets along with codes for the purpose of reproducibility of your results. You can simply share the processed datasets and ask people to cite the original paper. To me such not collaboration is a very bad sign. I really hope you share the copy of data and codes to 1) show your results are real 2) respect the the research community and your promises to release data and codes after publication. I consider this as severe misconduct. Even simple preprocessing can impact the final results, how do you expect people to reproduce the results while you are not helping in sharing codes and data?

…

On Mon, Jul 13, 2020 at 4:48 PM Shiyu Chang ***@***.***> wrote: We recently found that the beer review dataset has been removed by the original authors (https://snap.stanford.edu/data/web-BeerAdvocate.html). Due to this problem, we cannot publish the dataset at this point. You might consider contacting the original authors for a copy of the dataset. This is also unexpected to us. We could help to generate indices that match our partition once you have access to the original dataset. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AODM7IZRTR2LNL435HSFNALR3MNEDANCNFSM4OYA3KTA> .

Answer 5 · 2020-07-13T20:50:45.000Z

Please refer to #6