Data Mining

Major: Administration of Cybersecurity Systems
Code of subject: 7.125.04.O.001
Credits: 4.00
Department: Information Security
Lecturer: Khoma V.
Semester: 1 семестр
Mode of study: денна
Мета вивчення дисципліни: The purpose of teaching the discipline is to acquaint students with the tasks of intellectual data analysis, the principles of analytical research of large masses of information in order to identify new previously unknown regularities and useful knowledge necessary for decision-making; an overview of the various tools used for data analysis.
Завдання: KZ 5. Ability to communicate with representatives of other professional groups of different levels (with experts from other fields of knowledge / types of economic activity)
Learning outcomes: 1. Communicate freely in national and foreign languages, orally and in writing, to present and discuss the results of research and innovation, ensure business/operational processes and issues of professional activity in the field of information security and/or cyber security. 5. Critically consider the problems of information security and/or cyber security, including at the interdisciplinary and interdisciplinary level, in particular on the basis of understanding the new results of engineering and physical and mathematical sciences, as well as the development of technologies for creating and using specialized software. 14. To analyze, develop and support the system of auditing and monitoring the effectiveness of the functioning of information systems and technologies, business/operational processes in the field of information and/or cyber security as a whole. 16. Make well-founded decisions on organizational and technical issues of information security and/or cyber security in complex and unpredictable conditions, including using modern methods and means of optimization, forecasting and decision-making. 20. Set and solve complex applied engineering and scientific problems of information security and/or cyber security, taking into account the requirements of domestic and international standards and best practices.
Required prior and related subjects: Computer methods of high-level design of protection devices Security systems of intelligent objects
Summary of the subject: The discipline is devoted to the principles, models and methods of intellectual data analysis, which can be used to solve a wide range of tasks in various fields of science and technology. The discipline arose and is developing on the basis of such sciences as applied statistics, machine learning, database theory, etc.
Опис: Content and tasks of the discipline. The essence of intelligent data analysis (IAD), areas and tasks to be solved. Basic concepts in the field of intelligent data analysis. Data, information, knowledge. Data types, presentation formats and attributes. Data Mining and Big Data. Statistical methods of data analysis. Primary working results. Descriptive statistics. Statistical parameters and estimates of distribution moments of different order. Monte Carlo method Correlation and regression analysis. Correlation analysis of two and many variables. Regression analysis. The method of least squares. Method of support vectors. Principles of intelligent data analysis. Stages of intelligent data analysis. Data preparation and initial processing. The main tasks of IOD: classification, clustering, association, forecasting, evaluation. Solving the classification problem. Setting the problem of classification and presentation of results. Methods of construction of classification rules. Application of decision trees, support vector methods, "nearest neighbor", naive Bayes. Evaluation of classification errors. Solving the problem of clustering. Formal formulation of the clustering problem. Types of clusters. Presentation of results. Basic clustering algorithms. Proximity measures based on distances. Evaluation of clustering quality. Solving the association problem. Statement of the association problem. Templates and samples. Characteristics of associative rules. Methods of finding associative rules. Solving the forecasting problem. Formulation of the forecasting problem. Comparison of forecasting and classification problems. Forecasting and time series. Neural networks and their learning algorithms. Single-layer and multi-layer neural networks. A neural network model with back propagation. Using machine learning for natural language processing. Content and stages of natural language processing. Text classification, text extraction, automatic text and speaker recognition, machine translation, natural language generation.
Assessment methods and criteria: Current control of classroom classes is carried out with the aim of: • identifying the level of students' knowledge before starting classes; • ongoing verification of mastery of each studied topic; • assessment of the student's activity in the process of performing laboratory work; • verification of the execution and content of laboratory work reports; • inspection of control works. The final control is carried out based on the results of the test control and oral survey.
Критерії оцінювання результатів навчання: Performing laboratory work and protecting reports: - DFN (35) - ZFN (21) Performance of control works: - DFN (15) - ZFN (29) Final semester control of learning the theoretical course (45 points written and 5 points oral components)
Recommended books: 1. Основи теорії і практики інтелектуального аналізу даних у сфері кібербезпеки : навчальний посібник / Д. В. Ланде, І. Ю. Субач, Ю. Є. Бояринова. – К. : ІСЗЗІ КПІ ім. Ігоря Сікорського, 2018. – 297 с. 2. Інтелектуальний аналіз даних. Частина 1 / М.В. Талах, В.В. Дворжак – Чернівці: Технодрук, 2022. – 367 с 3. Wes McKinney. Python for Data Analysis, 3rd Edition. O'Reilly Media, 2022, 579 p.