data mining

With the rapid development of information technology, which is widely used in all spheres of human life and activity, extremely large amounts of data have been accumulated today. By applying machine learning methods to this data, new practically useful knowledge can be obtained. The main goal of this paper is to study different machine learning methods for solving the classification problem and compare their efficiency and accuracy.

Розглянуто класифікацію повідомлень груп новин у просторі семантичних полів. Проаналізовано ефективність баєсівського класифікатора та класифікатора за найближчими сусідами для різних навчальних та тестових вибірок повідомлень. Показано існування підмножини груп новин, для яких використання аналізованих класифікаторів є ефективним.

Проаналізовано проблеми підвищення точності нейроподібних структур машини геометричних перетворень (МГП) під час розв’язання завдань передбачення в галузі інтелектуального аналізу даних (ІАД). Розроблений в статті метод передбачення на основі виділення головних компонент шляхом об’єднання вхідних тренувальних та тестових даних ілюструється прикладом.

An analytical review of the development of Data Lakes and its application in various industries, as part of Big data concept solutions, was conducted. The available standard architectural solutions for the Data Lake organization are considered. Also, specialized areas that require different or additional aspects to solve the tasks, depending on the field of Data Lake use, are taken into account. For the proper organization of Data Lake, various data processing tools are used, including distributed data storage systems, semantic networks, and especially metadata.

Prevalence of diabetes in Gulf countries is knowing a significant increase because of various risk factors, such as: obesity, unhealthy diet, physical inactivity and smoking.  The aim of our proposed study is to use Data Mining and Data Analysis tools in order to determine different risk factors of the development of Type 2 diabetes mellitus (T2DM) in Gulf countries, from Gulf COAST dataset.

Досліджено особливості інтелектуальних компонент інтегрованих автоматизо- ваних систем управління. Розглянуто створення інтелектуальної компоненти ІАСУ для енергетичних систем. Запропоновано нейромережні методи прогнозування споживання електроенергії підприємством на основі машини геометричних перетворень. Наведено результати проведених експериментів.

This article describes the features of classification methods and technologies, analytics Big data. Described group of methods and technologies, analytics Big data that are graded according to the functional relationships and formal model of information technology. The problem of the definition of ontology concepts analytics Big data.

New information, telecommunication technologies contribute to the optimization in the management of studies. This paper is devoted to the implementation of innovative approaches to improving the curriculum of higher education. The method of finding and attracting students including girls for scientific and practical work through and their participation at team competitions and joint Interuniversity scientific-practical projects are proposed. This work has considered a problem of distance education and involvement in her adaptive learning system.

This article analyzes the current state of the models and methods of building recommendation systems. The basic classes of problems that solve the recommendation system are highlighted. The features of the method collaborative filtering are shown. Developed a method for calculating the similarity coefficients, taking into account the sparseness of ratings vectors of goods and people.

The paper analyzes the current state of development and application of recommendation systems, models and methods of construction of recommendation systems. It is shown that the most widely used method came into collaborative filtering. The method of fuzzy clustering is developed, which improves the accuracy of predicting ratings of products.