Serendip is an independent site partnering with faculty at multiple colleges and universities around the world. Happy exploring!

Week 9 (emergence)

Submitted by Paul Grobstein on Mon, 03/26/2007 - 2:51pm

Emergence 362

Your current thoughts about what you want to work on for your final modelling project?

Add new comment

i think i'd like to do a

Submitted by shikha on Mon, 04/02/2007 - 11:05am.

i think i'd like to do a model on whether consciousness is required to reach a tipping point. maybe i can make a model in which the agents are not conscious and another one in which they are - although i'm not quite sure how i would do that.

i'm also intriged by the rule of 150, so i might want to model that - perhaps try to see whether individuals change their opinions based on what the majority of the group thinks in a small grp vs a large grp, if the individuals are randomly assigned one of two opinions in the beginning.

Semester project

Submitted by mgupta on Sun, 04/01/2007 - 2:34pm.

Since I enjoyed reading Linked, I want to work on a project along the lines of links and nodes. Specifically, I want to work on how different situations disturb and alter the networks. Since I found the parallel with corruption networks really interesting, maybe modelling different situations in corruption networks would be really interesting. Still have to think how I would go about it, however.

Uh... I'm thinking...

Submitted by asmoser on Wed, 03/28/2007 - 2:50pm.

There are a couple different projects I've thought about which would be interesting. I might work to extend the inference model Frank and I made so that there are more turtles and they breed and die (generations). Not quite sure what I'd learn from that. To build on this, I think it would be really interesting to have some sort of genetic trait that would help with solving some specific problem. Whether this would be learning the most useful index for starting a partitioning algorithm or something completely different I don't know.

I'd also be interested in modeling the model for word of mouth epidemics that Malcolm Gladwell outlined in The Tipping Point. No idea what I'll actually do.

Scale-free networks

Submitted by Jessica B on Wed, 03/28/2007 - 1:31pm.

A lot of people have mentioned doing a project with some aspect of scale-free networks and I'm thinking of going this way, too. I'm not sure exactly how but a good first step might be to write a program that creates a network of nodes and links based on growth, preferential attachment, and fitness. Then I could toy around with it some and see what else I want to do. I'm not sure.

Modeling the Rise and Fall of Cities

Submitted by samkaplan on Wed, 03/28/2007 - 1:07pm.

I would like to model the ways in which different areas of cities thrive or decline. This will probably involve some modification of the game of life, probably with a lot more variables. I don't know. I have to think about somehow incorporating agents into the model, as well, to describe the effect that populations have on this process...

Still up in the air

Submitted by Lauren on Wed, 03/28/2007 - 12:37pm.

I have been going back and forth in terms of narrowing down project ideas, and though I have not settle on any one idea, I know that I want to build a model that either (a) extends this idea of emerging synchrony in group of objects that exhibit chaotic behavior or (b) simulates the ripple effect in small world network chains.

I know that many models already exist in the NetLogo library in regards to both these notions (i.e. fireflies, spread of disease, etc.), so like Heather, it's going to take a bit more brainstorming on my part to come up with something original. On the flipside, there are a lot of emergent behaviors mentioned in Strogatz's Sync that are interesting and may make excellent models, if only I could figure out how to translate them into NetLogo. Out of that "small" list of examples I presented last Wednesday, maybe something will jump out at me.

Just trying to pick a goal!

Submitted by biophile on Wed, 03/28/2007 - 12:19pm.

I have a couple of ideas, but they're quite vague and I'm not sure how to write them in net logo. Two things that I read about fascinated me: the synchronization of biological oscillators and small-world networks. I could incorporate both of them but I want to keep it simple for now. I saw some good models in the library: fireflies, AIDS, dining philosophers (which seems too complicated for my tastes), small worlds, preferential attachment, etc. It's just difficult, trying to think of a simple way to incorporate the more interesting parts of them into my own model.

Maybe I could have a bunch of nodes in a circle (the number could be changed on a slide bar) and have x number of them be circles and the rest triangles. The circles could spread the message to turn into circles to its neighbors. The connections could be changed from run to run, so you could see how the message is transmitted. Maybe there could even be different rules, i.e. if you're already a circle and you receive a circle signal you could turn into yet another shape...? It's quite hard to picture in my head and I'm sure it's been done to death. Or perhaps I could have different groups or pairs with different variables (frequency, position, etc.) communicating and responding to each other, while plotting the variables to see what exactly is going on. This is going to take a while...

I'd also be interested in...

Submitted by natsu on Sat, 03/31/2007 - 4:21pm.

doing something like your idea of "if you're already a circle and you receive a circle signal you could turn into yet another shape". That is, I'd be interested in making a model in which the agent can change due to some influence from its environment. But can agents change in Net Logo? I don't think I've ever seen an agent change itself in any model so far... It would be really cool if we can do that though...

I think Sync has set us on

Submitted by Lauren on Wed, 03/28/2007 - 12:39pm.

I think Sync has set us on very similar paths. :-)

Modeling ideas for a revolution-like phenomenon

Submitted by natsu on Tue, 03/27/2007 - 12:02am.

So I just wanted to list a bunch of ideas, most of which I'm not at all sure how to actually do with net logo.

I want to set up a world in which almost every single agent is acting in the same way, but without really being aware of it. Then, I want one (or maybe more than one, but very few) agent(s) to become aware of its unconscious thoughts, and start reporting what it noticed to the rest of the world. If possible, I want to also have what the agent is reporting to influence the other agents. Basically, my idea is to end up with a kind of revolution in which new a phenomenon emerge due to the fact that a few agents started thinking about and reporting something that the others had not yet been aware of. Yes, I realize that this is probably way too much to do, but I would really like to model at least one of these things...