Modern methods of data mining

Course unit code

В.1.2

Course unit title

Options

Name(s), surname(s) and title of lecturer(s)

Evgeny Pchelintsev, PhD, Associate Professor

Semester

3

ECTS credits

3

Working hours

Contact hours

lectures

32

labs

32

Self-study

44

Total

108

Work placement

Laboratory works in Computer class

Prerequisites

It is assumed that the students have mastered the following disciplines «Mathematical Analysis», «Linear Algebra», «Probability Theory and Mathematical Statistics», «Differential Equations», «Numerical Methods».

Language of instruction

English (Russian)

Objectives of the course

Learning outcomes

A student’s assessments methods

- This course is intended for training Masters in Math to apply mathematical methods and modeling technique in their professional work.

- Prepare a Master of Mathematics for applying statistical methods and numerical experiment for applications in professional activities.

- Connect theory and practice, to teach students to "see" the statistical problems in various subject areas and correctly apply the methods of applied statistics, practical examples show the possibilities and limitations of statistical methods.

In mastering the subject of Modern methods of data mining the student will acquire the following knowledge:

- the basic methods of applied statistics, the appointment, capabilities and features of modern statistical software (R, Statistica, etc.), advantages of the data processing using statistical software in comparison with other methods (tabular processors, database, programming, etc.).

To be able to:

- understand the posed problems,

- reasonably choose the data processing method according to the task and the type of data available,

- To carry out data selection, the various types of modifications and data conversion

- To freely navigate in the menu of the package and be able to use the various functions

- Present the results of treatment in the form of tables and graphs

- Use the syntax and create a data processing program,

- Competently draw conclusions and interpret the results obtained in the course of treatment with a statistical software.

The current control of mastering the discipline includes two written test and four reports on the labs.

The final control – exam.

Teaching methods

Lectures, Labs

List of Topics

Topic title

Contact hours

Assignments and independent study hours

Introduction. The primary data analysis

2

Parametric hypothesis testing

8

Written test 1

Non-parametric hypothesis testing

6

Lab 1

Correlation analysis

6

Regression analysis

8

Lab 2

Secondary data analysis methods

6

Multiple Hypothesis Testing

8

Lab 3

Time series analysis

8

Survival Analysis

6

Lab 4

Panel data analysis

6

Written test 2

64

exam

Assessment requirements

In during the semester 60 points

Assessment criteria

Each lab 10 points and test 10 points

The composition of final accumulative mark

Exam 40 points.

Examination ticket consists of two theoretical questions (10x2=20) and two exercises (10x2=20).

Author of the course

Evgeny Pchelintsev