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 & Processing| id |
drspiprikievua-article-239209 |
|---|---|
| record_format |
ojs |
| spelling |
drspiprikievua-article-2392092021-08-31T12:26:55Z A method of formation and clusterization of correlation networks of concepts Метод формування та кластеризації кореляційних мереж понять Ланде, Д. В. Страшной, Л. Балагура, І. В. кореляційна мережа, динаміка публікацій, GoogleBooksNgramViewer, візуалізація мережевих структур, кластерний аналіз cross-correlation network, publication dynamics, Google Books Ngram Viewer, visualization of network structures, cluster analysis 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. Для вирішення задачі формування та кластеризації понять запропоновано методику формування, кластеризації, ранжирування та подальшої візуалізації спрямованих кореляційних мереж, зв’язки яких визначаються на основі рядів динаміки, що відповідають цим поняттям. Як приклади розглянуто часові ряди динаміки вживання термінів, що формуються сервісом Google Books Ngram Viewer для формування кореляційної мережі наукових понять, і часові ряди динаміки захворюваності на коронавірус у різних країнах для формування та кластерізації мережі країн за ознакою подібності відповідних статистичних рядів. Наведена методика може застосовуватися з метою узагальнення множини сутностей без явно виражених зв’язків між ними на основі даних, отриманих в аналітичних системах різного призначення Інститут проблем реєстрації інформації НАН України 2021-06-29 Article Article application/pdf http://drsp.ipri.kiev.ua/article/view/239209 10.35681/1560-9189.2021.23.2.239209 Data Recording, Storage & Processing; Vol. 23 No. 2 (2021); 27-36 Регистрация, хранение и обработка данных; Том 23 № 2 (2021); 27-36 Реєстрація, зберігання і обробка даних; Том 23 № 2 (2021); 27-36 1560-9189 uk http://drsp.ipri.kiev.ua/article/view/239209/237860 Авторське право (c) 2021 Реєстрація, зберігання і обробка даних |
| institution |
Data Recording, Storage & Processing |
| baseUrl_str |
|
| datestamp_date |
2021-08-31T12:26:55Z |
| collection |
OJS |
| language |
Ukrainian |
| topic |
cross-correlation network publication dynamics Google Books Ngram Viewer visualization of network structures cluster analysis |
| spellingShingle |
cross-correlation network publication dynamics Google Books Ngram Viewer visualization of network structures cluster analysis Ланде, Д. В. Страшной, Л. Балагура, І. В. A method of formation and clusterization of correlation networks of concepts |
| topic_facet |
кореляційна мережа динаміка публікацій GoogleBooksNgramViewer візуалізація мережевих структур кластерний аналіз cross-correlation network publication dynamics Google Books Ngram Viewer visualization of network structures cluster analysis |
| format |
Article |
| author |
Ланде, Д. В. Страшной, Л. Балагура, І. В. |
| author_facet |
Ланде, Д. В. Страшной, Л. Балагура, І. В. |
| author_sort |
Ланде, Д. В. |
| title |
A method of formation and clusterization of correlation networks of concepts |
| title_short |
A method of formation and clusterization of correlation networks of concepts |
| title_full |
A method of formation and clusterization of correlation networks of concepts |
| title_fullStr |
A method of formation and clusterization of correlation networks of concepts |
| title_full_unstemmed |
A method of formation and clusterization of correlation networks of concepts |
| title_sort |
method of formation and clusterization of correlation networks of concepts |
| title_alt |
Метод формування та кластеризації кореляційних мереж понять |
| description |
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. |
| publisher |
Інститут проблем реєстрації інформації НАН України |
| publishDate |
2021 |
| url |
http://drsp.ipri.kiev.ua/article/view/239209 |
| work_keys_str_mv |
AT landedv amethodofformationandclusterizationofcorrelationnetworksofconcepts AT strašnojl amethodofformationandclusterizationofcorrelationnetworksofconcepts AT balaguraív amethodofformationandclusterizationofcorrelationnetworksofconcepts AT landedv metodformuvannâtaklasterizacííkorelâcíjnihmerežponâtʹ AT strašnojl metodformuvannâtaklasterizacííkorelâcíjnihmerežponâtʹ AT balaguraív metodformuvannâtaklasterizacííkorelâcíjnihmerežponâtʹ AT landedv methodofformationandclusterizationofcorrelationnetworksofconcepts AT strašnojl methodofformationandclusterizationofcorrelationnetworksofconcepts AT balaguraív methodofformationandclusterizationofcorrelationnetworksofconcepts |
| first_indexed |
2025-07-17T10:58:22Z |
| last_indexed |
2025-07-17T10:58:22Z |
| _version_ |
1850411525320212480 |