An implementation of social network group classification model based on recurrent and convolution neural networks

Authors

  • Denis M. Obolensky Sevastopol State University
  • Victoria I. Shevchenko Sevastopol State University
  • Olga V. Chengar Sevastopol State University
  • Elena N. Maschenko Sevastopol State University
  • Anastasia S. Soina Sevastopol State University

DOI:

https://doi.org/10.52575/2687-0932-2021-48-1-100-115

Keywords:

social networks, convolutional neural networks, recurrent neural networks, Python, Keras

Abstract

In this article, the authors consider the problem of classifying communities in the social network VKontakte. The application of neural networks for the classification of user groups according to the degree of radicality is investigated. In the work, a model of a recurrent long short term memory (LSTM) neural network is built using modern software technologies and methodologies. The resulting model is trained on a test dataset and is also evaluated using the selected metrics, such as F1, accuracy and loss. A convolutional neural network model is also built and evaluated. Subsequently, these models are compared with each other. The model which is based on convolutional neural networks had the higher value of metrics, than one based on the LSTM model. Methods for preprocessing data, as well as using the Keras framework for Python with Tensorflow backend for building neural network classifiers are proposed. The resulting convolutional network model is applicable to the core dataset when searching for radical communities on social networks. The research of the model is carried out and the analysis of the results is presented.

Downloads

Download data is not yet available.

Author Biographies

Denis M. Obolensky, Sevastopol State University

Assistant of the Department of Information Technologies and Computer Systems,

Sevastopol State University, Sevastopol, Russia

Victoria I. Shevchenko, Sevastopol State University

Candidate of technical sciences, Associate Professor, Head of the Base Department of Corporate Information Systems Sevastopol State University, Sevastopol, Russia

Olga V. Chengar, Sevastopol State University

Candidate of technical sciences, Associate Professor, Assitance Professor of the Department of Information Technologies and Computer Systems, Sevastopol State University,

Sevastopol, Russia

Elena N. Maschenko, Sevastopol State University

Candidate of technical sciences, Associate Professor, Assitance Professor of the Department of Information Technologies and Computer Systems, Sevastopol State University

Sevastopol, Russia

Anastasia S. Soina, Sevastopol State University

Candidate of technical sciences, Sevastopol State University, Sevastopol, Russia

References

Абрамов Р.А. Основные метрики задач классификации в машинном обучении. URL: https://webiomed.ai/blog/osnovnye-metriki-zadach-klassifikatsii-v-mashinnom-obuchenii/ (дата обращения: 12.08.2020).

Багутдинов Р.А., Саргсян Н.А., Красноплахтыч М.А. 2020. Аналитика, инструменты и интеллектуальный анализ больших разнородных и разномасштабных данных. Экономика. Информатика. 47 (4): 792–802. DOI 10.18413/2687-0932-2020-47-4- 792-802.

Батура Т.В. 2017. Методы автоматической классификации текстов. Программные продукты и системы. 1 (30): 85–99. DOI: 10.15827/0236-235X.030.1.085-099.

Котельников Е.В. 2012. Комбинированный метод автоматического определения тональности текста. Программные продукты и системы. 3 (109): 189–195.

Котельников Е.В., Клековкина М.В. 2011. Автоматический анализ тональности текстов на основе методов машинного обучения. РОМИП. URL: http://www.dialog-21.ru/media/1380/105.pdf. (дата обращения: 12.08.2020).

Кузьмицкий Н.Н. 2013. Создание универсальных классификаторов текстовых образов на основе сверточных нейросетевых технологий. ГрафиКон'2013: 23-я Международная конференция по компьютерной графике и зрению: 234–238.

Мащенко Е.Н, Оболенский Д.М., Соина А.С., Ченгарь О.В. 2020. «Разобщающий», «нейтральный» и «объединяющий» контент сообществ сети ВКонтакте на Украине: определение понятий. The Newman in Foreign Policy. 54 (98): 38–43.

Мошкин В.С., Андреев И.А. 2019. Сравнение эффективности применения алгоритмов сентимент-анализа неструктурированных ресурсов социальных сетей. Восьмая Междунар. конф. «Системный анализ и информационные технологии». САИТ-2019: тр. конф. М.: ФИЦ ИУ РАН: 534–540.

Метрики в задачах машинного обучения. URL: https://habr.com/ru/company/ods/blog/ 328372/ (дата обращения: 12.08.2020).

Оболенский Д.М. и др. 2020. Сбор и предварительная обработка данных пользователей социальных сетей с помощью портала «Открытые Данные». Modern Science. 7 (2): 369–378.

Рубцова Ю.В. 2015. Построение корпуса текстов для настройки тонового классификатора. Программные продукты и системы. 1 (109): 72–78. DOI:10.15827/0236-235X.109.072-078.

Смирнова О.С., Шишков В.В. 2016. Выбор топологии нейронных сетей и их применение для классификации коротких текстов. International Journal of Open Information Technologies. 4 (8): 50–54.

Шевелев О.Г., Петраков А.В. 2006. Классификация текстов с помощью деревьев решений и нейронных сетей прямого распространения. Вестник Томского государственного университета. 290 (1): 300–307.

Espíndola, Rogério & Ebecken, Nelson. 2005. On extending f-measure and g-mean metrics to multi-class problems. Sixth international conference on data mining, text mining and their business applications. 35: 25-34. URL: https://www.researchgate.net/publication/286914533_On_extending_

f-measure_and_g-mean_metrics_to_multi-class_problems (дата обращения: 12 августа 2020).

Jo T. 2010. NTC (Neural Text Categorizer): Neural Network for Text Categorization. International Journal of Information Studies. 2(2): 83–96.

Kolluri J., Razia S., Nayak S.R. 2020. Text Classification Using Machine Learning and Deep Learning Models. URL: https://ssrn.com/abstract=3618895 (дата обращения: 12 августа 2020).

Kowsari J. M., Heidarysafa M. 2019. Text Classification Algorithms: A Survey. Information, 10 (4): 150–154.

Maas A.L., Daly R.E., Pham P.T., Huang D., Ng A.Y., Potts C. 2011. Learning word vectors for sentiment analysis. The International Language Technologies. 1: 142–150.

Nguyen T.H., Shirai K. 2013. Text Classification of Technical Papers Based on Text Segmentation. Natural Language Processing and Information Systems. NLDB 2013. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, 121.

Ramasundaram S., Victor. S. 2010. Text Categorization by Backpropagation Network. International Journal of Computer Applications. 8 (6): 1–5.

Zhang, X. 2015. Character-level convolutional networks for text classification. Xiang Zhang, Junbo Zhao, Yann LeCun. In Advances in Neural Information Processing Systems. 2015. Feb. 649–657 p.

Understanding LSTM Networks. URL: http://colah.github.io/posts/2015-08-Understanding-LSTMs (дата обращения: 17 July 2020).

Nguyen T.H., Shirai K. 2013. Text Classification of Technical Papers Based on Text Segmentation. In: Métais E., Meziane F., Saraee M., Sugumaran V., Vadera S. (eds) Natural Language Processing and Information Systems. NLDB 2013. Lecture Notes in Computer Science, vol. 7934. Springer, Berlin, Heidelberg.

Ramasundaram, S., Victor, S. Text Categorization by Backpropagation Network. International Journal of Computer Applications. 2010. 8 (6): 1–5.

Zhang, X. Character-level convolutional networks for text classification. Xiang Zhang, Junbo Zhao, Yann LeCun. In Advances in Neural Information Processing Systems. 2015. Feb. 649–657 p.

Understanding LSTM Networks. URL: http://colah.github.io/posts/2015-08-Understanding-LSTMs (accessed: 17 July 2020).


Abstract views: 70

Share

Published

2022-09-19

How to Cite

Obolensky, D. M., Shevchenko, V. I., Chengar, O. V., Maschenko, E. N., & Soina, A. S. (2022). An implementation of social network group classification model based on recurrent and convolution neural networks. Economics. Information Technologies, 48(1), 100-115. https://doi.org/10.52575/2687-0932-2021-48-1-100-115

Issue

Section

COMPUTER SIMULATION HISTORY