Machine learning step by step occupies an ever-increasing place in our world due to the increasing number of its applications, widening day by day. You can begin with the analysis of traffic jams and end with self-driving cars, people address a big amount of tasks to self-learning machines. So let’s explain the main idea of machine learning for beginners.
ML is considered to be a part of more common and detailed concept of AI, the main idea of which is that the computer does not just use a pre-written path called algorithm, but comprehends how to solve the problem itself.
Any working machine learning technology can be conditionally attributed to one of three levels of accessibility. The first level is when it suits only to various technological giants of the level of Google or IBM. The second level is when a student with a certain amount of practice in programming can use it. The third level is when even your granny can use it simply.
Machine learning is now at the junction of the second and third levels, due to which the rate of change of the world with the help of this technology is growing every day. What are the machine learning steps in general?
TRAINING: WITH OR WITHOUT A TEACHER
The tasks of machine learning can be divided into learning with a teacher (supervised learning) and training without a tutor(unsupervised learning). By “teacher/tutor” we mean the very idea of human intervention in the data processing. When training with a teacher, we have data on which to predict something, and some hypotheses. When teaching without a teacher, we only have the data, the properties of which we want to find.
All tasks that are performed during machine learning can be divided into:
- The task of regression: on the basis of various signs to predict a real answer.
- The task of classification: on the basis of various signs to predict a categorical answer.
- The task of clustering: splitting data into similar categories.
- The task of reducing the dimension: learn to describe our data not with N signs, but with a smaller number (as a rule, 2-3 for subsequent visualization)
- The task of identifying anomalies: on the basis of signs, learn to distinguish anomalies from “non-anomalies.”
Another important concept in machine learning for beginners are neural networks.
This is one of the universal algorithms, but those who start working with machine learning very vaguely represent the principle of their work. A neural network (or artificial neural network) is a network of neurons, where each neuron is a mathematical model of a real neuron being a math function in fact. With the help of neural networks, you can solve at least the regression and classification problems and build extremely complex models.
And if in practice, how to explain the principle of machine learning to beginners?
A simple example: machine learning was used by a team of data analysts to prepare President Obama’s election campaign: by solving simple tasks, they determined the president’s target audience and predicted what time and state he should hold meetings with voters to raise the rating.
And a second example: the famous targeted advertising. The features built into the site’s engine, with your visit and the interest of any product, will then aggressively advertise this or similar product to you everywhere. And this, too, is machine learning.
So, having the main idea of what machine learning is, what should you do to dive into it deeply?
If you are one of the machine learning beginners you should at least understand mathematical analysis, linear algebra and optimization methods (with an emphasis on the last two) for in-depth study of the subject with an understanding of the entire foundation. In addition, it is desirable to know the basics of programming and any programming language. In machine learning, R, Python or Matlab are usually used.