Citizen Science Grid

The Citizen Science Grid is run by Travis Desell, an Assistant Professor in the Computer Science Department at the University of North Dakota. It is hosted by UND's Computational Research Center and Information Technology Systems and Services. The Citizen Science Grid is dedicated to supporting a wide range of research and educational projects using volunteer computing and citizen science, which you can read about and visit below.

Climate Tweets

The Climate Tweets project is focused on personal opinions about climate change or global warming. The goal is to sort tweets and view the different views in various countries, how the discussion has changed over time, and how opinions change with political orientation. Classifying tweets allows us to discover patterns and coorelations in people's opinions about our world. It also helps us understand what people know about climate change. Please note that the tweets are unfiltered and may contain profanity or controversial views, and these are not the views of the Citizen Science Grid, any of our team, or funding agencies. Because of this the project is 18+.


Wildlife@Home is citizen science project aimed at analyzing video gathered from various cameras recording wildlife. Currently the project is looking at video of sharp-tailed grouse, Tympanuchus phasianellus, and two federally protected species, interior least terns, Sternula antillarum, and piping plovers, Charadruis melodus to examine their nesting habits and ecology.


The Subset Sum problem is described as follows: given a set of positive integers S and a target sum t, is there a subset of S whose sum is t? It is one of the well-know, so-called "hard" problems in computing. It's actually a very simple problem computationally, and the computer program to solve it is not extremely complicated. What's hard about it is the running time – all known exact algorithms have running time that is proportional to an exponential function of the number of elements in the set (for worst-case instances of the problem).


The goal of DNA@Home is to discover what regulates the genes in DNA. Ever notice that skin cells are different from a muscle cells, which are different from a bone cells, even though all these cells have every gene in your genome? That's because not all genes are "on" all the time. Depending on the cell type and what the cell is trying to do at any given moment, only a subset of the genes are used, and the remainder are shut off. DNA@home uses statistical algorithms to unlock the key to this differential regulation, using your volunteered computers.

User of the Day

User profile
Profile Tobas
..the tendecy is to push it as far as you can..

News RSS

[wildlife] image credit now being exported to badges.xml

Image credit is now being exported to as <image_credit>. This should allow signature generators, etc to use the correct image_credit images.

Interface enhancements to the image interface are still being tweaked, but I'm hopeful all will be implemented by the weekend and ready for testing.


Marshall Mattingly on Monday, January 9th
leave a comment

[wildlife] credit and fpops update

Looks like my estimates for fpops (and credit) were quite a bit low. I've increased them by a factor of 6, which I think should be more in line with what we should be awarding, with a little added bonus for putting up with the new app while it's in alpha testing.

Let me know how this works out. All new workunits should be awarded the new amount of credit and have the new FPOPS estimate.

Travis Desell on Wednesday, January 11th
leave a comment

[wildlife] testing exact assimilator

So far results coming in for the v0.11 app look really good. Not seeing any conflicts as of yet, which is good news!

I've also fired up the assimilator and am watching it manually. It should be automatically generating new workunits as work starts flowing in. Let me know if you have any problesm!

Travis Desell on Saturday, January 7th
leave a comment

[wildlife] v0.11 apps

I've updated the applications to version 0.11. I believe I've fixed all inconsistencies so checkpointing and validation should really be working correctly now. It took me a bit of time as I had to write a few things by hand involving random number generation that weren't working across operating systems and then double (or triple check) that everything was running the same everyone.

On my Windows, Linux and OS X systems I'm getting the same results, so I hope this should fix those issues so I can get back to updating the assimilator and getting a steady flow of work going!

Let me know if you're having any issues.

Travis Desell on Sunday, January 15th
leave a comment

[wildlife] update on validation issues

Progress update for today. I've found the issue with validation. I *am* generating the same stream of random numbers from the random number generator I'm using from C++'s standard library on linux and the other platforms.

The bug ended up being that the standard library shuffle operation I'm using to randomly shuffle the training images is implemented differently on linux vs OSX vs windows, which is resulting in the different output.

Looks like I need to implement my own random shuffle algorithm so I know it's doing the same thing across all the OSes. Strange thing was that originally I was getting the same answers on all of them (maybe an update to the Linux OS changed the shuffle implementation).

Should have updated apps out tomorrow which should put this issue to rest!

The assimilator is also almost complete, which will let me generate a steady stream of workunits.

Travis Desell on Sunday, January 8th
leave a comment