View unanswered posts | View active topics It is currently Tue Mar 19, 2024 1:16 pm



← Back to the Calcudoku puzzle page




Reply to topic  [ 43 posts ]  Go to page 1, 2, 3, 4, 5  Next
 what to do with the data? 
Author Message
User avatar

Posted on: Thu Nov 10, 2011 7:23 pm




Posts: 3295
Joined: Thu May 12, 2011 11:58 pm
 what to do with the data?
[confused] [confused] [confused]

So now the milestone of 1 million solved puzzles has been reached,
what to do with the data?

For every solved puzzle I have the obvious information (when it was solved,
size, level, etc., the time taken for timed puzzles), and also:
- country where the solution was submitted from
- browser and operating system used

Any suggestions for interesting information to be extracted from all this data??

Patrick


Profile
User avatar

Posted on: Fri Nov 11, 2011 7:14 pm




Posts: 3295
Joined: Thu May 12, 2011 11:58 pm
Post Re: what to do with the data?
A first easy one, for example: the total number of points scored for those 1 million puzzles was about 3.5 million..

Patrick


Profile

Posted on: Fri Nov 11, 2011 8:28 pm




Posts: 855
Joined: Fri May 13, 2011 6:51 pm
Post Re: what to do with the data?
pnm wrote:
A first easy one, for example: the total number of points scored for those 1 million puzzles was about 3.5 million..

Patrick


Hi, Patrick, my input:

Total points by country
Total points by continent
Total points first year
Total points second year
Average points by day of the week
Average points by month of the year
Global average (all players) timed puzzle 4x4
Global average (all players) timed puzzle 5x5
Global average (all players) timed puzzle 6x6
Less number of solvers for a 9x9 (and date it happened)
Less number of solvers for a 10x10 (and date it happened)
Less number of solvers for a 12x12 (and date it happened)
...


Profile

Posted on: Sat Nov 12, 2011 12:11 am




Posts: 246
Location: Lisbon, Portugal
Joined: Sun Sep 18, 2011 5:40 pm
Post Re: what to do with the data?
Reading the message of Clm I thought that a good parameter is also the two first sugestions of Clm but also, per capita.

_________________
Visit http://www.calcudoku.org the most interesting and addictive site of puzzles.


Profile

Posted on: Sat Nov 12, 2011 1:26 pm




Posts: 49
Joined: Sun Oct 16, 2011 10:58 am
Post Re: what to do with the data?
Graphs.
For example:
- x-axis date (month-year) / y-axis total in that month
- color = puzzle type


Profile

Posted on: Mon Nov 14, 2011 5:12 am




Posts: 54
Joined: Thu Nov 03, 2011 8:52 am
Post Re: what to do with the data?
One interesting statistic would be to know the average number of times that puzzles of each size were solved. It would be easy to obtain it by dividing the number of times 4x4 puzzles were solved by the number of 4x4 puzzles published; and then do the same for 6x6 and the other sizes. It would give you an idea of how popular the different sizes are compared to each other. You would probably have enough data to do it on a per-month or per-quarter period. That would tell you whether with time people "graduate" to more difficult puzzles.

I guess, you didn't really need to wait for 1 Mpuzzles to do that. I would have started making statistics from day one. Haven't you? [biggrin]

Tell us what you already have!


Profile
User avatar

Posted on: Wed Feb 29, 2012 1:09 pm




Posts: 3295
Joined: Thu May 12, 2011 11:58 pm
Post Re: what to do with the data?
pnm wrote:
So now the milestone of 1 million solved puzzles has been reached,
what to do with the data?

Alright, as you saw on the main page, I've done something with the usage data... :-)

This took a _long_ time to produce (most of my spare time in the past two months),
I'll be happy to go back to programming on the site again.. [mellow]

Patrick


Profile

Posted on: Wed Feb 29, 2012 4:18 pm




Posts: 116
Joined: Sat May 14, 2011 3:18 am
Post Re: what to do with the data?
I've just read your paper on "Choice of Browser and Numerical Intelligence," and I think your conclusion that Chrome users have the highest numerical intelligence is based on incomplete data. That is, you only had the raw times and browser type on which to base your conclusion. But given that you found no statistical difference between browser types for the non-timed puzzle and the statement made elsewhere on this forum that younger solvers do better on the timed puzzles, I think if you had access to the age of each solver you would find that a more appropriate conclusion is that "Chrome users are typically younger than IE (and Firefox) users."

The older a person is, the more likely they are to stick with what they know and are comfortable with. Older people are (on average) also less technically inclined than the new generation that can't remember a time before they had access to a personal computer and the Internet.

So while it's possibly true that Chrome users are more technically adept on average than IE users, I think age is a far better indicator than browser type.

Cheers,
Pharosian -- whose primary browser is Safari


Profile

Posted on: Thu Mar 01, 2012 12:05 am




Posts: 212
Joined: Fri May 13, 2011 2:11 am
Post Re: what to do with the data?
Having read the article as well, I'm partially wondering how much this would differ from a study that found that, say, Native born French speaking people are more likely to use Firefox and the like.

That is, I'm beginning to think of puzzle speed as a language, rather as math is considered a language, etc.

First of all, a short version of this, for people who don't want to read the long story below: I've been doing puzzles my whole life, and so when I started doing kenkens, it was the same as speaking English to me, and that, more than anything else, is why I feel I can do them quickly relatively speaking.

For me, one of my oldest memories (Which, technically, isn't very long ago. A little under 12 years, but still) is of a trip my family took to Europe, when I had just turned 7. I don't remember much of that trip at all. I remember going to Rodin's Gardens, solely because of the Thinker; I remember the Eiffel Tower because when it lights up it's so pretty; I remember Mount Pilatus, and how beautiful Lucerne was; I remember the Trevi Fountain, and the Galleria Dell'Academia, and the canals and stained glass of Venice.

But as much as any of that, I remember what I did on the trains between cities; I worked puzzles. Nothing difficult, no Kakuros with a 45 row, no 16x16 Kenkens, etc. More along the lines of http://www.jimloy.com/puzz/mensa.htm. Just very basic number puzzles.

Then, fast forward to 6th grade (At which point I would have been 11), at which point I was still doing number puzzles, though the math involved had gotten slightly heavier. I started Middle School, and for the first time got to participate in Math Team. Math Team, for clarification, is effectively a bunch of number puzzles, generally more complex than the earlier ones, but still not immensely complex.
The highest level of it, as far as I'm aware, is the type of competition that this site sprung from; Limited by my school's travel resources, I never got up to that level, though I was at least competitive against the people who did take the USAMO and were at the national level, and that was only in terms of difficulty. In terms of pure speed, I was faster than any of them because all of their knowledge had been taught, whereas all of mine had been learned. As a result, at the two speed oriented competitions, I beat them every single time I knew how to do the problem.

Meanwhile, while still doing math team, since math team runs middle school through high school (When I was 11 to when I was 18), when I was 14, my dad started keeping a Kakuro book with him in his car. I had done a few (Probably around 500 or so, don't underestimate this number) Sudoku in the past, but had worn out on them because they're extremely limited as to what you can do with them. I still work them from time to time, but I just don't find them entertaining enough. So from when I was 14 to about when I turned 16, my dad and I would occasionally just take some free time to work a Kakuro puzzle. This usually was when it wasn't useful to drive home after church and drive back, and the like.

When I turned 16, my Dad and I no longer had reason to spend that time working Kakuro together, since I could drive to church on my own. So I started working more of them on my own.

Then, of course, I discovered this site as I was getting bored with Kakuro, and I've been doing Kenken's ever since.

All this time, of course, corresponds to the approximate timing of my critical period (en.wikipedia.org/wiki/Critical_period). Basically, in a large percentage of cases, people who learn second languages during this time frame are more likely to gain a fluent state.

Similarly, sheldolina, as she was 13 as of the last time she mentioned her age, likely learned how to do puzzles during her critical period.

As a result, the two of us seem to speak the language fluently, as opposed to just recognizing phrases as they come at us.

But basically, to me, puzzles like this don't register as a puzzle, it registers as a language. It's the same concept as http://www.chessblog.com/2012/01/classi ... rtise.html, where chess grandmasters see a board as a sentence; They have no trouble repeating a sentence back to you, but a random set of words is difficult.

I remember a little while back, someone saying that a little thing you could do with these puzzles was try and turn the monitor off and do them. This is the same type thing as the chess experiment, and it comes fairly natural to me for easier puzzles, because typically, it's like remembering a sentence. However, for the very easiest ones, the ones that are addition only, it becomes more difficult -- almost like remembering a sentence like "I am tired and so I will sleep." Every word makes sense, but none of the words are interesting, and so it becomes harder to remember precisely than a sentence with more complex thoughts in it.

As for the actual article, as a Firefox user, I'm actually surprised to see that the gap between Firefox and IE is so small.

Also, oddly, I think that helps the argument that Chrome's success over Firefox is not simply the result of ages of solvers, since a least 1/2 of the two examples of age helping people excel at this use Firefox.

So yeah, to anyone who read all the way through this, sorry for rambling a bit; I just had an odd realization about how the puzzle language works.


Profile

Posted on: Thu Mar 01, 2012 2:02 am




Posts: 98
Joined: Fri May 13, 2011 12:48 am
Post Re: what to do with the data?
-> starling. I'm 37 and have very similar memories on doing puzzles that you have. Yet I still fall far far behind both you sheldolina. I think that the age factor is more significant than you think(as well as that of typing speed)

Although... sometimes I know the right path 'intuitivly' even as I am stuck logically. Maybe I have just lost the language?


Profile
Display posts from previous:  Sort by  
Reply to topic   [ 43 posts ]  Go to page 1, 2, 3, 4, 5  Next

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
cron
All forum contents © Patrick Min, and by the post authors.

Forum software phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by STSoftware.