updated mnemosyne dataset

[lars klein]

Hi,

I would like to do some research on spaced repetition and your dataset looks amazing.

The last version comes from 2014 though, would it be possible to get an updated version?

There was a thread about this in 2016, where Peter said he could upload to FTP.

Does that offer still stand ?
I could mail you a link to my server :)

I would also try to make this available over torrent for other interested users but can't make any promises about uptime.

In any case, thanks for making this awesome project and supporting research.

The 2014 dataset will be a huge help.

Cheers,

Lars

[Peter Bienstman]

Hi,

 

If you have e.g. an ftp site where I could drop a few gigs, I could do this, provided that if you do serious academic research on this data leading to a publication, I get involved as a co-author. The reason as that knowing how the data is collected allows to prevent certain pitfalls, where other publications have fallen into.

 

This also means I’d rather not have you make the data available on a torrent.

 

Cheers,

 

Peter

--
You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.
To post to this group, send email to mnemosyne-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/1ef2e71f-9bda-4b3a-8351-58f84d266b41%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[lars klein]

Hi Peter,

thanks for the quick answer.

This dataset would (hopefully) motivate some design choices in a larger project.

I'm working on this as my masters thesis. Therefore, involving you as a co-author is not my decision to make.

Sharing our analysis and results with you is certainly possible and feedback would be very welcome.

But they would be a small component in a much larger overarching project.

I don't know the common practice for support like this, maybe there could be a "special thanks" section in the publication (if it is published).

Kind regards,

Lars

On Wednesday, November 28, 2018 at 10:59:12 AM UTC+1, Peter Bienstman wrote:

Hi,

 

If you have e.g. an ftp site where I could drop a few gigs, I could do this, provided that if you do serious academic research on this data leading to a publication, I get involved as a co-author. The reason as that knowing how the data is collected allows to prevent certain pitfalls, where other publications have fallen into.

 

This also means I’d rather not have you make the data available on a torrent.

 

Cheers,

 

Peter

 

From: mnemosyne-...@googlegroups.com <mnemosyne-...@googlegroups.com> On Behalf Of lars klein
Sent: 28 November 2018 10:27
To: mnemosyne-proj-users <mnemosyne-...@googlegroups.com>
Subject: [mnemosyne-proj-users] updated mnemosyne dataset

 

Hi,

I would like to do some research on spaced repetition and your dataset looks amazing.

The last version comes from 2014 though, would it be possible to get an updated version?

 

There was a thread about this in 2016, where Peter said he could upload to FTP.

Does that offer still stand ?
I could mail you a link to my server :)

 

I would also try to make this available over torrent for other interested users but can't make any promises about uptime.

 

In any case, thanks for making this awesome project and supporting research.

The 2014 dataset will be a huge help.

Cheers,

Lars

--
You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-users+unsub...@googlegroups.com.
To post to this group, send email to mnemosyne...@googlegroups.com.

[Peter Bienstman]

Hi,

 

Feel free to have your supervisor contact me off-list J

To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.

To post to this group, send email to mnemosyne...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/1ef2e71f-9bda-4b3a-8351-58f84d266b41%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.
To post to this group, send email to mnemosyne-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/28e3b43a-408f-478d-b9fc-02b198387a9d%40googlegroups.com.

[Gwern Branwen]

Are you sure you need an updated dataset? The 2014 one is enormous
already and it's difficult to see what you would do with another few
million rows that you couldn't do with the previous millions of rows.

--
gwern
https://www.gwern.net

[lars klein]

Hi Gwern,

you're surely right.

My request was more a matter of principle.

In a previous project I made the mistake of changing the dataset while fitting models.

Which invaluated previous results.

Since I don't know what preprocessing and cleaning is necessary, it made sense to try and get the most data possible before starting to work.

Kind regards,

Lars

[lars klein]

Hi,

I would like to follow up on this.
The master's project is mostly done - and has taken a turn in a different direction. So a publication based on our ideas plus your dataset is sadly no longer on the table.

However, now I have the time to tie up some loose ends. And I still find spaced repetition rather interesting.
A close friend of mine speaks lithuanian. And I would like to experiment with a custom language training app for that language.

For this project, I would really love to bootstrap from your mnemosyne data.
Would it be possible to receive the logfiles for this purpose?

This is not intended to lead to a paper. The scope are a few hacky weekends.
That being said, in the unlikely event that it transforms into a scientific publication somehow, having you as a co-author would not be a problem.
Not to mislead you though, it is not my intention to publish this.

What do you think?
kind regards,
Lars

[Peter Bienstman]

Hi,

Just contact me privately.

Cheers,

Peter

-----Original Message-----
From: mnemosyne-...@googlegroups.com <mnemosyne-...@googlegroups.com> On Behalf Of lars klein
Sent: 22 April 2019 14:50
To: mnemosyne-proj-users <mnemosyne-...@googlegroups.com>
Subject: Re: [mnemosyne-proj-users] updated mnemosyne dataset

--
You received this message because you are subscribed to the Google Groups "mnemosyne-proj-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.
To post to this group, send email to mnemosyne-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/1fd4433d-9911-475b-9b80-8e8eb9c2a497%40googlegroups.com.