A method of formation and clusterization of correlation networks of concepts
A method for forming, clustering, and visualizing correlation networks is herein proposed. The links between nodes of such networks correspond to the values of cross-correlations between vectors — sets of parameters corresponding to these nodes modified in a certain way. To build network structures...
Збережено в:
Дата: | 2021 |
---|---|
Автори: | , , |
Формат: | Стаття |
Мова: | Ukrainian |
Опубліковано: |
Інститут проблем реєстрації інформації НАН України
2021
|
Теми: | |
Онлайн доступ: | http://drsp.ipri.kiev.ua/article/view/239209 |
Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
Назва журналу: | Data Recording, Storage & Processing |
Репозитарії
Data Recording, Storage & ProcessingРезюме: | A method for forming, clustering, and visualizing correlation networks is herein proposed. The links between nodes of such networks correspond to the values of cross-correlations between vectors — sets of parameters corresponding to these nodes modified in a certain way. To build network structures for each node (topic), vectors are formed — arrays of numbers corresponding to a certain time series. The research considers the time series of the dynamics of the use of terms as examples: 1) formed by the Google Books Ngram Viewer service for the formation of a correlation network of scientific concepts; 2) time series of the dynamics of the incidence of COVID-19 in different countries for the formation and clustering of the network of countries, based on the similarity of the relevant statistical series. To build a network of concepts related to modern trends in Computer Science, data obtained by accessing the Google Books Ngram Viewer service was considered as an information source. The idea is to cluster topics with similar movements to identify trends in science. As an example, we consider 20 concepts. In the second example of health data as a result of the analysis of 50 countries, the corresponding correlation matrix was obtained, a network was formed and its clustering was carried out. This technique can be used to generalize a set of entities without explicit links between them based on data obtained in analytical systems for various purposes. Examples of subjects that you can apply the presented method:1) political leaders, parties characterized by their attitude to various spheres of public life; 2) consumers of products — the parameters here are sellers and the sources of products; 3) entities and concepts reflected in social media, in this case, parameters can be time-series of published volumes for certain time periods. Tabl.: 1. Fig.: 10. Refs: 9 titles. |
---|