Have you had that experience when it seems like your laptop is reading your thoughts? When Netflix offers you exactly those shows or movies that resonate with you and meet your interests. Or when you are browsing your favorite online store and a personalized offer with a suitable product pops up at you. Although it might seem the big brother is watching you, machine learning is what stands behind all these coincidences.
There is no doubt that machine learning is a prospective and constantly developing field. But for most of us, it would be difficult to decide where to start and what to learn first. To make your life easier, we have prepared a list of 5 outstanding resources to learn and practice ML on your own. All of them are free which makes it easier for beginners to get basic practical skills and acquire their first experience.
Basic Things to Know About Machine Learning
Machine learning is an application of artificial intelligence which allows a computer to draw conclusions not in a direct way, but based on a search for the patterns. The purpose of ML is to automate processes and solve certain problems using the algorithms that can imitate the work of a human brain when making decisions. These algorithms are capable of finding the patterns in huge amounts of data and as a result you get the right solution for the certain task.
To run through the basics, there are 2 main types of machine learning algorithms:
Supervised learning is used to teach the machine to recognize certain objects or signals and the programmer controls the process. This method requires a complete set of data so the machine knows exactly which patterns to look for. Supervised learning is generally used for regression and classification. It is suitable for tasks when there is a large set of accurate data for algorithms.
Unsupervised learning is used when the machine doesn’t know the right answer and what patterns it should look for. The machine still has a dataset but there are no leads what to do with it. By analyzing data, the network tries to look for all patterns it can find. Such algorithms can reveal similarities and differences. Unsupervised learning is suitable for digital marketing, clustering, finding associations and fraud detection.

Where to Use Machine Learning?
Studying machine learning is a good decision as ML algorithms are currently applied across a variety of domains. Here are just a few of them:
- Social Media. The news feed is based on the content that users comment and like regularly. If the system is sure that you will like the certain post, it will show you similar content more often. For instance, if you comment, like and save every post of Ryan Reynolds, after a while his posts will be the first to appear in your news feed.
- Security. Face recognition systems can scan people’s faces and compare them with photos of wanted criminals. If the system detects a high level of similarity, it gives a signal for law enforcement.
- Hotels. When you get ready for a trip and try to choose a good hotel browsing booking sites, you’ll definitely get recommendations that would suit your needs, budget and location. Recommendations based on user’s preferences, comparison of hotels, recognition of fake reviews – this is only a part of what machine learning is capable of.
- Medicine. Collection and processing of patients’ data, diagnosis of diseases, prognosis and selection of treatment.
Top-5 Free Resources for Machine Learning
- Andrew Ng’s ML Course
You will need to spend about 54 hours to complete an introduction to ML. The course includes supervised and unsupervised learning, various case studies, database mining, theoretical and practical tasks to apply your knowledge at once. Course materials are free, thus you only need to pay if you want the certificate and graded assignments.
2. Google’s Machine Learning Crash Course
This course is brief compared to the previous one (about 15 hours), but it focuses on the practical tasks, real case studies and exercises. Take this course when you understand the basic theory in order to concentrate on the practical tasks and get your first results.
3. Kaggle
If you want to immerse yourself into ML, you will need data to learn and analyze patterns for any of your tasks. Here you can find over 19 000 public datasets and lots of real ML problems to solve. Whether your goal is to predict prices or you want to forecast how many people would survive the ship crash, ML will make a perfect match for you. These and many other practical tasks are available in the “Competition” section on Kaggle.
4. Microsoft Azure Machine Learning
Modern cloud services allow you to build, train, deploy your own ML models and explore various frameworks and tools. Microsoft Azure ML is constantly updated and new features are added so you are always aware of the latest changes.
5. IBM Watson Machine Learning
Another helpful cloud service to put your own ML models into production is IBM Watson Machine Learning. One-click deployment, dynamic retraining, model operations and other necessary features are available to run your ML models and come at no cost. The resource is completely free so you don’t need to fill in with your credentials or any other personal information.
Machine learning is a rather sophisticated and rapidly developing field. But if you are ready to study hard and pursue your goal every day, you have a great chance to build a successful career. Your success is heavily dependent on your persistence and determination, so everything is possible!