Introduction to Machine Learning Machine Learning: An Introduction to Basic Concepts and Applications Introduction: This week, we will introduce you to the basic concepts in Machine Learning, and show you some example notebooks using Scikit-learn, a popular Machine Learning library in Python.You won't be an expert in Machine Learning after this week, but you will be able to gothrough and understand Machine Learning analysis, the basic ones, as well as create simpleMachine Learning tasks in Python. You will build upon this knowledge and know more aboutMachine Learning after taking the Machine Learning class in this Micro Masters. So take thisweek as an appetizer to what's coming later. What is Machine Learning? Let's start by defining machine learning as it is the topic of this week. Nowadays, the phrase "machine learning" is used frequently and in a variety of circumstances, so it'simportant to establish a clear definition right once. Computer systems that can learn fromdata are the subject of the topic of study known as machine learning. In other words, thesesystems—often referred to as models—can learn to carry out a given task by examiningseveral examples of a certain problem. For example, a Machine Learning model can learn to recognize an image of a cat by being shown lots and lots of images of cats. This notion of learning from data means that aMachine Learning model can learn a specific task on its own. The Machine Learningalgorithm is programmed to learn from the data that there's nothing in the algorithm orprogram which directly aims to learn the given task. In other words, Machine Learningmodels are not given the step by step instructions on how to recognize the image of a cat.Instead, the model learns on its own what features are important in determining that apicture contains a cat from the data that it has analyzed. The Importance of Data for Machine Learning: It's important to note that since the model learns to perform this work from data, the quantity and caliber of data that were made accessible for developing the model are crucialcomponents in how well the model learns from the task. As machine learning models areable to learn from data, they may also be used to find hidden patterns and trends. Whenusing machine learning, these trends and patterns provide insightful information about thedata, enabling the creation of data-driven functions for a specific issue. Applications of Machine Learning in Everyday Life: Machine Learning has been used in many different applications, many of which you probably encounter in your everyday life, perhaps without even realizing it. One application
of Machine Learning that you likely used this past weekend, or even just today, is credit cardfraud detection. Every time you use your credit card, the current purchase is analyzedagainst your history of credit card transactions to determine if the current purchase is alegitimate transaction or a potentially fraudulent one. If the purchase is very different fromyour past purchases, such as for a big-ticket item in a category that you never show aninterest in, or at the point of sale location is from another country, then it will be flagged as asuspicious activity. In that case, the transaction may be denied or you might get a phone callfrom your credit card company to confirm that the purchase was indeed made by you. This isa very common use of Machine Learning that's encountered in everyday life. Handwritten digit recognition is another practical example of machine learning at action. The numbers on a handwritten check are read by a machine learning algorithm when youdeposit it into an ATM. The technology analyzes each digit's characteristics—such as theheight, width, and curvature of the strokes—using image processing techniques beforecomparing them to a database of recognized handwritten digits. The check can beprocessed and the appropriate amount credited to your account once the system hasidentified the digits on the check.