So essentially, we've now reached the end 
of the course. But before I leave you,

I wanted to speculate on three trends and 
necessities of agent-based modelling

that I see are vitally important as we 
move forward with agent-based modelling.

And this is my particular take, these are
things I'm passionately interested in,

but it's also capitalizing on trends we're
seeing in other areas of computational

sciences, especially computational social
sciences and biological-socio-ecological

sciences. And one of them is the automatic
generation of agent rules. Now, this is

something I talked about in 9.1, right? 
We have this huge amount of big data

that's now being collected about what 
humans and entities are doing around the

world, right? We have sensors on 
everything, we have the internet of things

exploding left and right. We have social
data, we have app data, we have data about

the phones in everybody's pockets, right?
And the question is, can we use that

in some way to define rules automatically
that capture the essence of human actions

in those spaces, right? Like define how
humans move through the world, define how

humans interact with the internet of
things, right? Or define how the internet

of things interacts with itself, right,
as cars become more and more automated.

Of course, these rules also need to be 
validated, right? And one of the ways we

can do the validation is by building up
these rules, making predictions

with them seeing these predictions
carried out.

Now why would it be interesting to do this
at an agent level, right?

If we have all this data, why can't we 
just use the data, right?

But without the agent rules that predict
individual-level behavior,

we don't really have the ability to assess
what would happen if we were to change

the incentive structure for one particular
individual in that space.

Now causal state modelling gives us one
example of this.

But we could use many others, 
you could use decision trees,

you could use the associative rules, right.

You could use classifier systems. There are 
a lot of methods

out there that allow you to do this.

And with all these sources of data, big data,

administrative data is not new, but the ability to process it in large amounts is,

natural language data, text to speech data,

social data, app data, really being able to

capitalize on this will powerfully change the way

agent-based modelling is perceived in the world around us, right.

And give us a much more powerful tool kit 
to address some of those situations.

Now, of course, one of the big problems with some of this

data and one of the problems that people
often complain about, right is that

trace data is essentially just digital exhaust.

It is data that doesn't tell you anything.

So we need ways to validate those models

against real-world data and calibrate

those models in order to show that they're
actually working well, right.

We need rigorous guidelines, I believe, to follow,

to show that our models, the agent-based 
models, have been validated appropriately.

I often think of this as a statistics-like 
suite of tests, right.

Statistics is very good. The discipline of
statistics is very good.

at saying, "If you have data looks like 
this and your outputs look like this,

then these are the various tests 
that you need to apply," right.

And in some cases, all we would be doing is

basically discovering which of those
appropriate statistical tests, that had

already existed, would be most likely to 
apply in that particular situation, right.

And then, what that means is that lends
a lot of credence and credibility to

the idea that this model I created and
run through these tests to compare to

empirical data that I've seen have now then, have helped me to validate and

increase my confidence in this model.

Now, of course, being able to calibrate our models

in order to increase that level of validation
would also be useful.

So making tools like BehaviorSearch, which
we discussed earlier, easier to use

so users can calibrate models automatically against

real-world data would be a very powerful step forward in this space.

Finally, and this kind of stems off of
both of those other thoughts about

the future, if we could build models
automatically... if we could construct

them from these vast amounts of data that

are coming in - and we could continuously
validate them in some automatic way

where we have a suite of tests that
we know will tell us whether or not

the model is behaving accurately, then we

theoretically could build a tool which
will automatically construct a model

on the basis of streaming data, right?

and by that I mean it will be pulling
down... for instance...

stock-ticker data, Twitter data, whatever

it needs, and continually making a 
model of, say, the socioeconomic status

of the country as a whole, for instance, 
or something like that.

Or maybe it's pulling in data from the
internet of things, that looks at

sensors that are attached to all of the
street lights in a particular city

that detects the traffic flow, and then
making continual predictions

about whether or not there's going to be

traffic jams in certain areas

so that policy makers and city managers
can, in real time, change signage

in order to have people move around 
traffic jams.

If that was the case... if we had this
ability, this tool to build these

frameworks in real time that were 
predictive, well, not predictive,

but able to forecast and explore scenarios
into the future,

we could then use this to support
real-time decision-making.

So those are three areas - the use of
streaming data, big data

and the ability to automatically create
rules from big data,

and this validation and calibration
and improvement, that I think are

critical for the future of agent-based
modelling.

Like I said, that's it. Unit 9, we talked
about big data ABM, we talked about

design guidelines. We talked about 
the uses of ABM for communication

and education. We talked about advanced
programming constructs like map and

reduce and run and run-result task.

We talked about participatory simulations,
systems dynamic modelling

then we talked about extensions and the
future of agent-based modelling.

You'll see the Unit 9 slides, and you'll
see the Unit 9 tests coming up shortly.

But I want to thank everyone for
participating in this course.

I hope you enjoyed it. I enjoyed
teaching it for sure,

and I've really enjoyed a lot of

the discussions that we've been having

in the forums, and through the YouTube
office hours, and things like that.

And I hope that you are able to gain
something out of this.

Please, stay in touch as you go forward

and use agent-based modelling in your

own work or for your own fun, 
for that matter.