In this module, we focus on a very relevant type of data, and how to visualize this type of data. We're talking about temporal data. Very important type. So, what is temporal data? I would say, generally speaking, temporal data is data in which the values that you see in the dataset depend on time. At the same time, time is recorded as one or multiple variables in your dataset. Just to give you an example, here in this table, we have a dataset of vehicle collisions in New York City, and every single row represents one collision. As you can see, there are two columns that are related to time. There is a date column, and there is a time column. Now, date and time represent the date and time when these vehicle collisions happen. So, since we have data, and all the values that you have in this table depend on this temporal information, we can say that this is a temporal dataset. So, what is really important and interesting about temporal data and visualization, is that it's really ubiquitous. There are so many problems and datasets out there, that are that can be described as temporal data analysis and communication problems. Actually, if you think about it, virtually every single dataset is implicitly or explicitly some time, because data must have been recorded at a certain time. Right? So, I would say, in general, the main difference is that some datasets record time explicitly, whereas in some other datasets, time is just not a valuable thing. But if you think about it again, time is always there. Data must have been collected at a certain time. So, examples of domains where there is plenty of temporal data are for instances, businesses, in datasets that describe natural phenomena, or datasets that describe behaviors, think about human behavior, or animal behavior. Think about animal moving across the globe and being tracked, or cars being tracked, and so on. So, traffic and mobility, of course, also medical and healthcare datasets, information about patients, how something some measurements change over time. Finance, of course, is very heavy on temporal data and so on. I could go on forever. So, there are two sub-types of temporal data that is important to keep in mind. The first one is event data. What is event data? Well, the idea in event data is that every single object in your dataset represents one event. So, typically, what you have is some tiny information, plus the attributes that describe this object. In general, you can identify these type of temporal data, if it can be described as something happened at time T. So, some examples of this kind of data, for instance, let's say, any message on social media would be an event. So, every single message, say, an email, or a tweet, or a message on Facebook, and so on. Or say for instance, in computer security, in cybersecurity, an alarm that has been triggered by a server, this is a single event that happens at a given time, and potentially with a number of attributes that describe this event. Another very important class that is different from event data, is measurements. Measurements that are linked to time. So typically, what you have is something that is measured at a specific time and date. So, say for instance, information coming from sensors, say temperature, or maybe in finance the revenue, or stock value of a given stock in the market, and so on. So typically, this is measurement data that is linked with time. So, one important thing to keep in mind here, is that you have two main types of temporal data, event data, and measurements.
