Data Science

Apply now

The biennial master’s program in English is devoted to research in the field of Data Science. In the modern digital world, more and more sought-after professionals have competencies in Data Science. The knowledge gained through this program will allow graduates to work in various companies around the world. In any branch of human activity, it is necessary to process data, find hidden patterns, and predict the future behavior of an object according to its past state. This is all a specialist area of expertise in Data Science.

The field of Data Science is huge. The curriculum is designed in such a way that graduates of our program gain knowledge and skills about various areas of Data Science. In addition to theoretical training, much attention is paid to practical exercises.

All teachers leading disciplines are highly qualified, have a degree not lower than a PhD and have extensive practical experience in their field of activity.

The educational process uses modern technology and teaching aids. Modern computer labs, multimedia lecture halls, classes for mathematical modeling, unlimited access to the digital library, modern E-Learning and distance learning technologies used in the process of mastering the program will help to successfully master this master’s program.

The master program is designed and implemented in accordance with the Federal State Educational Standard of Higher Education in the field of training 09.04.01 Computer Science & Engineering.

For the development of general cultural, general professional and professional competencies in the NUST MISIS created all the favorable conditions.

General information about the curriculum

The curriculum of the Master’s program focuses on the courses that are essential for the profession:

Natural and artificial intelligence

This course serves as introduction to the brain morphology and basic psychophysiology. We will talk about common patterns of human behavior: motivation, emotions, choice, learning and recognition. We also give an overview of brain mechanisms of our social behavior (altruism, xenophobia, empathy, theory of mind, friendship, sexuality and morality). The course consists of lections and discussions and will be useful for everybody interested in neuroscience.

Machine Learning

Discipline is devoted to the study of a number of sections of statistical analysis. Different classes of statistical models and methods for their construction will be considered. Students will need to acquire practical skills in solving problems of predictive analytics using statistical models and algorithms. Main sections are: Statistical Learning, Linear Regression, Moving Beyond Linearity, Classification, Unsupervised Learning, Support Vector Machine.

Modern storage and data processing hardware

The course covers modern technologies and hardware, used for storing and processing big arrays of data, with Huawei systems as an example. By the end of the course students will have knowledge about technological solutions used for effective and reliable storage of data and operations on them.

Network architecture and modern networking solutions for data science

The discipline is aimed to the theoretical knowledge and practice on network technologies, based on a mix of Cisco and Huawei “Routing and switching” licensed courses. Students who passed this course will be able to set up networking equipment, plan out multi-level networks and administer created networks.

Data collection and preprocessing using specialized Python libraries

This course covers essentials of data collection, filtration and basic data preprocessing. Students who finish this course will possess knowledge of several data anomaly detection methods, ways of data classification, sorting and cleaning. Students will be able to use SQL to collect data from hardware, and code Python based data processing algorithms, using Pandas, SciPy, openCV, Tensor Flow and several other specialized libraries.

Elements of statistics

Everything that deals even remotely with the collection, processing, interpretation, and presentation of data belongs to the domain of statistics, and so does the detailed planning that precedes all these activities. Indeed, statistics includes such diversified tasks as calculating the batting averages of baseball players; collecting and recording data on births, marriages, and deaths; evaluating the effectiveness of commercial products; and forecasting the weather. Even one of the most advanced branches of atomic physics goes by the name of quantum statistics.

The word “statistics” itself is used in various ways. It can be used, for example, to denote the mere tabulation of numerical data, as in reports of stock market transactions and in publications such as the Statistical Abstract of the United States or the World Almanac. It can also be used to denote the totality of methods that are employed in the collection, processing, and analysis of data, numerical and otherwise, and it is in this sense that “statistics” is used in the title of this course.

There are several reasons why the scope of statistics and the need to study statistics have grown enormously in the last seventy or so years. The main one is the increasingly quantitative approach employed in all the sciences, as well as in business and many other activities that directly affect our lives. This includes the use of mathematical techniques in the evaluation of antipollution controls, in inventory planning, in the analysis of traffic patterns, in the study of the effects of various kinds of medications, in the evaluation of teaching techniques, in the analysis of competitive behavior of business managers and governments, in the study of diet and longevity, and so forth. Also, the availability of computers has greatly increased our ability to deal with numerical information, so that sophisticated statistical work can be done even by small businesses and by college and high school students.

The subject of statistics can be presented at various levels of mathematical difficulty, and it may be directed toward applications in various fields of inquiry. For example, governments have long used censuses to count persons and property, and the problem of describing, summarizing, and analyzing census data has led to the development of methods that, until recently, constituted about all there was to the subject of statistics. These methods, which at first consisted primarily of presenting data in the form of tables and charts, make up what we now call descriptive statistics. This includes anything done to data that is designed to summarize, or describe, without going any further; that is, without attempting to infer anything that goes beyond the data themselves. Although descriptive statistics is an important branch of statistics and it continues to be widely used, statistical information usually arises from samples (from observations made on only part of a large set of items), and this means that its analysis requires generalizations that go beyond the data. As a result, the most important feature of the recent growth of statistics has been a shift in emphasis from methods that merely describe to methods that serve to make generalizations; that is, a shift in emphasis from descriptive statistics to the methods of statistical inference.

These two directions are included in the supposed course. In difference to the most accepted in Russian educational tradition to consider statistics as a part of probability theory and teach the subject as a mathematical one, in this course main attention is devoted to descriptive statistics and to meaningful aspects of statistical inference. It does not mean that mathematical and especially algorithmic questions are not included into the course; it means merely that mathematics plays subservient role to studying and using methods and models of modern statistics.

Artificial neural networks

Since the beginning of the development of artificial neural networks in the middle of the XX century, there has been increased interest in them. And such advantages as the ability to generalize, adapt and universality of the approach of information processing have made artificial neural networks one of the most popular data processing tools. With the development of computers and a significant increase in their performance, it became possible to create large neural networks and process a large amount of data. Since the beginning of the 10th years of the XXI century there has been an explosive development of deep neural networks that can accurately solve the tasks of pattern recognition, forecasting and others.

The course involves the study of the basics of artificial neural networks, as well as the most popular types of neural networks: perceptron, Kohonen self-organized maps, neocognitron. After that, it is proposed to study data preparation and the basics of designing deep neural networks. In the process of studying it is necessary to perform laboratory work on solving applied problems with the help of artificial neural networks. The project-oriented approach allows not only to study the design and functioning of neural networks, but also to solve other subtasks on which the solution of the applied problem will depend.


For support in registration, accommodations, or any questions about the admissions process, please contact with the International Master’s Programs Admission Office

119049, Russia, Moscow, Leninsky Prospect, 6, office G-366

007 499 230 27 97

For detailed information on the education program, please contact the program’s education studies advisor, Prof. Alexander