So now we want to go ahead and see can we actually build a model like the one behind that web tool that we show you and that you can now go and play with yourselves? The answer is yes. I think we can do that today and so to do that maybe let's first think a little bit of what is it that we want to do and maybe to put some specifics so we can actually think about a specific problem. So let's think about flu and let's think about a specific question. So again, we create a model to help us maybe predict how many people may get infected with the flu during one year, during one season with some parameters and then use that model to answer this question, how many people would we need to vaccinate before the outbreak starts in order to prevent it? So let's have that specific question in mind and then let's see if we can come up with a model. So you might ask yourself, so how can we do that? Well, this brings to the whole concept of computational thinking, right? We have a complex problem that in this case transmission of the flu throughout the population and how can we deal with that, maybe simplify it and transform it into problem we can actually tackle and maybe use a computer or computers to help solve it. So what do we need to do is think about what are the key ingredients, what are the key issues, what are the key players that we need to think about and care about and then once we do that and identify what are the key things that can happen and then translate all of these into some rules, mathematical computational rules that we can then use or implement in a computer. So to sounds easy, right? So let's see if we can actually do it. So to begin with, let's think for a moment of how does a disease like the flu propagate across population, how does this happen? So while we're doing this, I asked you to think about your own places, your own settings, your own towns, cities and think about well, how does the flu propagate in my town, in my cities? If you live in the city maybe you start thinking, "Why go to the subway and then someone sneezes next to me and then maybe I'm going to get it, right? If you live in a small town maybe you think, "Well, if I go to school, that's usually when I meet some sick people and that maybe I interact with them and that's how I get it. So depending on where you live and settings might be where most infections may actually be happening so think about that, think about your own towns due to your own experiences. How does the flu propagate there? It actually probably goes from one person to person like someone is sick then they sneezes, they cough, they touch different surfaces and someone else either gets exposed to the sneeze or maybe touches the same surface and then they touch their eyes and then they get the virus into their own bodies. Where this is happening is mentioned maybe in subways, maybe in that image that you're seeing there, maybe it's at schools, maybe it's in that coffee shop, maybe where people work. So different places in this case people with infection will get or have contact with people who don't have the infection and then propagate it. How did the people get infected? So we talked about when someone cough, sneezes, touch surfaces so basically you need to get in touch with virus coming out from the sick or an infected individual and then get the virus into your own body. Then if your immune system is not able to prevent that infection from happening, then the virus will get into your system and they'll start propagating inside and then cause your own infection. So what are things that happen and when? What is the timeline if all of these things? So think about this and now that we have that mental picture in our minds maybe we can start thinking about well, what can we do in order to create a computational model to help us answer some of these questions. So now that you have that mental picture and some ideas of how the flu virus might be going around town infecting new people, let's now take all that complexity and really hold those details which are complicated and then every person is going to be affected in different ways at different times and boil down to maybe some simplified ingredients or key elements that are critical for this to happen. So maybe one way to help us think about it is to first think, "Okay. So who are the key players? Who is here? Who are the key actors? Who are the groups or agents that we want to consider?" Maybe you could start thinking, "Well, maybe it's the virus that's the main agent, the main player." That's a first thing. In this case, we're thinking about flu among humans so it's people. If we were thinking about maybe animals or some type of swine flu or avian flu, we maybe thinking about other animals but in this case we're thinking about people. So who are the key individuals that we care about? Well, we need to care about those who are sick and can pass the virus and maybe some who are healthy and actually can get the virus and continue the change of events. We also talk about well, maybe some of important players are the places where people will come together and facilitate the transmission of these buyers. The other piece now that we have some of who are the key players is also to think about well, what are the key things that can happen and what are the main things that can occur to in this particular case, these agents, people virus. Well, people can get infected it means the virus gets into their own bodies and then they develop their own infection. Usually, how is this happening? Well, there many biological processes that we could think about. But in very rough terms where someone who has the virus has to come in touch or in contact with someone who doesn't, and that's how it propagates, of course, there are many other things that happen behind the scenes biological individual, but in the end, in very rough terms, yes that's what has to happen. A sick person had to come in contact with someone who doesn't have or hasn't had the infection and that's how it happens. So they need to be in close contact. It could be in the same space and time, so they've meet each other and see each other or it could only be that someone like for instance I'm here, I touch my screen, and then the next person who's going to come and record the video touches the screen and then that happens. So maybe contact we can think about it in different ways, but at least there has to be some type of interaction either at same time or at least in the same space between two people. Then once that someone is infected, then what happens to that person? Well, usually, you get infected and eventually you start developing symptoms. Eventually you feel bad enough that maybe you want to stay at home, but before that you were going around. So they can actually, you could actually go and infect others. When you start getting some of those symptoms and you're at school during that day. So then you pass it some of your friends. But eventually you probably will maybe take a break or, eventually, You will just start getting better. So there are just things that can happen, right? Once you're infected you can pass the virus or you could recover as likely you're going probably be doing both, and once you're recovered then that's an important thing. Can you get it again or not. Usually, wtih some of these infectious systems. What happens is that when you recover, that means that your immune system develop some defenses against this virus, and now the next time you face the virus now, your body is going to be able to tackle it and prevent you from getting sick again. So now we say that you're immune. As well as you go with the vaccine later. So I think now that we have I guess some idea of who are the key players and what are things happening, we at this point usually need to make a decision. How much detail do we want to have in our model? Do we want to track every single virus particle? Do we want to track every infection, every individual, every contact, or do we want to do something simpler. Maybe just look at aggregate behaviors. So we need to think about those things and in some sense modelers and computational thinkers will have some debates about the best way of doing these. How close do you want to be to reality versus how much maybe abstract you want to be so you can have more control of things. There's a lot of approaches to these. Usually one thing that we want to think about is, well, what is a question that you have and then based on your question, and also maybe based on the time that you have, you might decide what different approaches to take. So in reality is in some ways how much, what is the specific questions. So you want to develop the model that allows you to answer that question. But maybe as complicated as you need to without being more than what you need to, but also will depend on how much time you have, and also how much data that you have to inform that model. So these are things that will leave us out today. For today what we're going to do is maybe go with something simple, more abstract, maybe less close to reality, but that can still help us think about some of these issues. So some decisions, let's focus on people. Let's not focus on what happens to every tiny virus, and then what we need to do if we're going to people then let's break them into a few classes and then go from there. So reflecting undertow, and just reflecting on what can happen to people. What people can be healthy, but be at risk of getting infected, people getting infected, and that point be able to transmit the virus and then eventually they clear it and they recovered. Then maybe later we'll do vaccination. So we'll maybe break then a population into some classes. Those who can get infected, susceptibles, those who have the virus and can infect others, and those who had the virus and now recovered and now they are immune. So we'll focus on those three groups, through classes. Then we're going to think about what can happen to those people in each of these different classes. Well, the susceptibles can get infected. So once they're infected, do sick people, infected people, can recover and become immune and they can transmit the virus. So those will be the main things. So now what we're going to do is then maybe start building a model by actually drawing a diagram, here show you that it's helpful, that it's easy for us to think about some of these processes or some of these questions were actually do a diagram or a drawing, right? So we said well we have three types of individuals. We said we have the susceptibles, let me call them Ss, and then we have maybe those who are currently infected or infectious and we're going to call them I, another box, and then we're going to have that recovers, and the point is that when you're here you could get infected when you're in the S. When you're in the I, you can infect someone or you can recover, and when you're in the R, I think now you're done. You're those who won't do anything else with respect to transmitting the flu. So in some sense, an I can become an R so we should maybe or we could put an arrow between the Is and the Rs and say, well, infected people might recover, and then we can think about well and how they recover, how quickly they recover. So for now let's just say well, infected people are infectious people or whatever you want to call them are going to recover at a certain rate or we're going to pull up parameter there right? There is going to thrive. How the Is are becoming Rs. The Is that's what they can do, the Rs remains Rs, they're immune, they cannot do anything anymore. But the Ss can get infected, right? So we know that there might be some of the Ss who as they getting contact with the Is, they might get infected. So we can put an arrow between the Ss and the Is, and then we can think again that maybe there's a rate or a parameter that regulates how the Ss are going to be becoming Is as they get in touch with each other. So this a very simple diagram, but it will help us summarize everything that we said, right? It shows us the three classes of agents. The Ss, the Is, and the Rs and the two main things that can happen. Infected people get in touch with Ss and then those Ss becomes I, and then the infected can actually recover and then become an R, capital R. So these is basically the model that we could create and start working with. A model that has three classes, three compartments people will call it, and that there's started three. The two things that I can actually happen in the model, right? The susceptibles becoming infected, infected recovering, and then the question is, can we actually now go ahead and translate these into an algorithm that we could put in a computer which will be the focus of the next optional video.
