Hello. Welcome to module 16, lesson one.
This lesson we'll introduce you to the concept of anomaly detection.
More familiarly, you might have heard about this
as outlier detection or even fraud detection.
The idea is can we find something that's different than the rest of the data i.e.
an anomaly and can we do it in an automated way that makes it
easy to do so consistently across a large data set.
So in this lesson, I want to introduce you to the concept of what an anomaly is and
how different approaches might be used to find them and why it might even be important.
So specifically, by the end of this lesson,
you should be able to describe how to identify an anomaly,
articulate different types of anomalies that might be present in
a data set and explain a way that accountants might encounter and identify anomalies.
There will be four readings for this particular lesson.
The first is three ways accountants can use Big Data to fight fraud.
This is a nice little article that talks about the fact that fraud is everywhere and it's
a very important component
particularly for people doing audit to be able to recognize it.
And one of the key things that I'm trying to do throughout this course is to encourage
you to think about not just today but your entire career.
Right now, we're seeing a revolution in data impacting all areas of society.
And one way to do that as a business person
is to be able to use that data to your advantage.
When we have lots of data,
it may make it easier in some senses to hide fraud via large amounts of data.
But on the other hand, if you know what you're doing you
can actually use that to be able to
recover issues that might be present in a large data set that might seem to be hiding.
So, this article does a nice job of talking about
all these different ways that you might think about finding fraud,
being able to identify why it might be happening,
and then how to use Big Data or just the data analytics that you're
learning in this course to be able to find examples of that.
The next article is a Wikipedia article,
it talks about data analytics techniques that might be used for fraud detection.
I like this because it's pretty readable and it actually gives you
lots of examples of where you might want to
try different techniques as well as some of the ways
different career opportunities are becoming present for doing the sorts of work.
So, this is a nice way of introducing you to use different techniques
both just how it's done as well as some of the machine learning techniques.
And you'll recognize many of the ways,
the techniques that they're saying to use this in order to find fraud.
The supervise and unsupervised learning techniques that we've learned in this course.
Next article is one from Deloitte,
talking about detecting fraud and how you can do this particularly in retail.
And this is a really interesting article that
talks about how one example of it was found,
how big the problem is and ways that things might have to
change in the future in order to minimize or reduce some of these problems that you see.
The last article talks about how machine learning can be
used in an online sense for retail and how to find fraud in a retail,
and talks about ways that companies can lose revenue and can thus be hurt if they're
not careful in not using fraud detection
to prevent the loss of revenue for their particular company.
So hopefully, I've given you a good introduction to
anomaly detection in different ways that it might be present,
in terms of online retail, regular retail,
typical accounting techniques, but also I hope that I've shown you that this
is a really important and growing topic
and one that you might want to explore more as your career unfolds.
If you have any questions, let us know and,
of course, good luck.