Научный журнал «Инновации. Наука. Образование» Индексация в ринц н Инновации. Наука. Образование
жүктеу/скачать 22,11 Mb. Pdf көрінісі
|
Номер 51 февраль 2022 года
| Инновации. Наука. Образование
Роза Виталий Петрович Роза Мария Петровна Студент магистратуры и аспирантка Сибирский Государственный Университет Имени академика М. Ф. Решетнева METHODS OF CLUSTER DATA ANALYSIS AND CLUSTERING QUALITY ASSESSMENT Аннотация: В данной статье определяется важность использования методов автоматической группировки, а также приводится оценка качества кластеризации. Ключевые слова: анализ данных, кластеризация, автоматическая группировка, кластеры. Keywords: data analysis, clustering, automatic grouping, clusters. Modern Russian researchers present automatic grouping problems in the form of integer linear programming problems. Currently, there are a huge number of different effective methods for solving optimization problems. It happens that difficulties arise during the solution of large tasks of automatic grouping, taking into account the fact that there is an instantaneous increase in the volume of data that is collected and processed in automated systems. Clustering, based on the established similarity relation of elements, establishes subsets (clusters) into which the input data is grouped. One of the simplest and most effective are methods and models based on minimizing the total distances between objects of the same group (cluster) or between cluster objects and its center. Automatic grouping methods are used in many branches of science, including actively used in data mining systems. Due to the use of models of optimal placement and automatic grouping of objects, the requirements for economic efficiency are increased. Automatic grouping methods can group objects by constructing models of the relationship of objects in a continuous space of characteristics. Such methods have the opportunity to be applicable with sufficiently large amounts of data. At the same time, tasks should be solved interactively with a limited working time with a large amount of input data. 816 Научный журнал «Инновации. Наука. Образование» Индексация в РИНЦ н жүктеу/скачать 22,11 Mb. Достарыңызбен бөлісу:
|