The course is part of these learning paths
Machine learning is a branch of artificial intelligence that deals with learning patterns and rules from training data. In this course from Cloud Academy, you will learn all about its structure and history. Its origins date back to the middle of the last century, but in the last decade, companies have taken advantage of the resource for their products. This revolution of machine learning has been enabled by three factors.
First, memory storage has become economic and accessible. Second, computing power has also become readily available. Third, sensors, phones, and web application have produced a lot of data which has contributed to training these machine learning models. This course will guide you to the basic principles, foundations, and best practices of machine learning. It is advisable to be able to understand and explain these basics before diving into deep learning and neural nets. This course is made up of 10 lectures and two accompanying exercises with solutions. This Cloud Academy course is part of the wider Data and Machine Learning learning path.
- Learn about the foundations and history of machine learning
- Learn and understand the principles of memory storage, computing power, and phone/web applications.
It is recommended to complete the Introduction to Data and Machine Learning course before taking this course.
The datasets and code used throughout this course can be found in the GitHub repo here.
Hello and welcome to this video on supervised learning. In this video, we will talk about supervised learning and some successful applications of supervised learning and we'll also discuss different types of supervised learning. I'd like to start with an example. Have you ever noticed that email spam is practically nonexistent? This is one of the great success of supervised learning. In the early 2000s, email boxes were plagued by a ton of emails advertising pills and other fraudulent activity, money-making schemes, and in general very useless information. The first step to get rid of these was to allow users to move spam emails into a spam folder. This action provided the training labels. Then with millions of users manually cataloging spam, large email providers like Google and Yahoo could quickly gather enough examples of what spam looks like to train a model that would predict the probability for a message to be spam.
That's an example of what's called a binary classifier. And this is a machine learning algorithm that learns to distinguish between two classes like true or false, dead or alive, positive or negative, or simply one and zero. Binary classifiers trained with supervised learning are present everywhere. You find them in predicting churn, for example, this is a very common application in telecom when you want to predict if a user is about to go to a competitor or not. It's a yes, no type question. And this model is useful to know when and to whom to make an offer for a new product. Another application is sentiment analysis. This is what a social media company would do if you are, for example, a celebrity that has millions of fans in its Facebook page or in social media page. This is what a social media company would do if you are a celebrity and have million of fans following your page. You'd post something and you will receive a million of comments and you want to quickly know whether or not what the average sentiment about your post was. If you are a celebrity, you receive millions of comments each time you post something on a social media page, and the question is how can you know if your followers were primarily happy or angry or perplexed about what you tweeted. A sentiment analysis classifiers can tell you that for each single comment and therefore you can get another reaction by aggregating the response. Other applications of binary classifiers include predicting whether a user is going to click or not on a link, screening the presence of a disease, detecting if you're a human or a bot, think of captures, and many more.
A natural extension of binary classifiers is the case where more than two classes are present in the labels. This is called a multi-class classification and neural networks can naturally deal with it. Supervised learning is also used to predict continuous quantities. For example, to forecast retail sales next month or to predict how many cars there will be at a certain intersection in order to offer a better route for your can navigation, predicting weather, predicting the value of a stock, or predicting a biosignal based on previous data that you've collected. In all these cases, the labels are not discrete, like true or false or black, blue, green, red, but they have continuous values like 21, 23.5, 18, and so on, whatever value they may take. So stop for a second and see if you can come up with at least a couple of examples of either discrete or continuous supervised problems that you are interested in. In this video, we've seen a few applications of supervised learning and we've talked about binary classification, multi-class classification, and regression for continuous values. Thank you for watching and see you in the next video.
I am a Data Science consultant and trainer. With Catalit I help companies acquire skills and knowledge in data science and harness machine learning and deep learning to reach their goals. With Data Weekends I train people in machine learning, deep learning and big data analytics. I served as lead instructor in Data Science at General Assembly and The Data Incubator and I was Chief Data Officer and co-founder at Spire, a Y-Combinator-backed startup that invented the first consumer wearable device capable of continuously tracking respiration and activity. I earned a joint PhD in biophysics at University of Padua and Université de Paris VI and graduated from Singularity University summer program of 2011.