The task of this master thesis is to review the field of machine learning and to review the basic steps of machine learning, its tools and methods. Also, find and analyze the set of data and conduct machine learning on it using the methods described. Machine learning is part of artificial intelligence whose task is to draw conclusions based on given data. Its purpose it to learn using different algorithms and models so that the acquired knowledge could be applied to the new data. With the increase in the amount of data in the world, machine learning is evolving more and more as are its methods. The two most important types of machine learning are supervised and unsupervised machine learning. One of the main types of supervised machine learning is classification, which uses examples with labels so that the computer can learn. The goal is to develop a model that successfully determines class affiliation for a given problem. A multiclass classification was used in this paper to show its meaning and contribution to machine learning in everyday life. The classification was performed using logistic regression and neural networks. Logistic regression is one of the most commonly used classification algorithms. It enables both binary and multiclass classification by fitting the model parameters and cost function. For good results, a well-set hypothesis is most important. Neural networks are a set of algorithms that allow the simulation of neurons in the brain, hence their name. They consist of an input, output and hidden layer. For good results, it is important to find optimal number of hidden layers, as well as the number of neurons and the activation functions that are appropriate for the task. An example of data classification was done in Python, and models were created using the sklearn library. For the purposes of this paper, a data set consisting of online traffic records was used and the data were classified by the type of connection for each example.