MULTILAYER ADAPTIVE FUZZY PROBABILISTIC NEURAL NETWORK IN CLASSIFICATION PROBLEMS OF TEXT DOCUMENTS

Authors

  • Ye. V. Bodyanskiy Kharkiv National University of Radioelectronics, Kharkiv, Ukraine, Ukraine
  • N. V. Ryabova Kharkiv National University of Radioelectronics, Kharkiv, Ukraine, Ukraine
  • O. V. Zolotukhin Kharkiv National University of Radioelectronics, Kharkiv, Ukraine, Ukraine

DOI:

https://doi.org/10.15588/1607-3274-2015-1-5

Keywords:

classification, adaptive fuzzy probabilistic neural network, overlapping classes, neurons in the data points.

Abstract

The problem of text documents classification based on fuzzy probabilistic neural network in real time mode is considered. A different
number of classes, which may include such documents, can be allocated in an array of text documents. It is assumed that the data classes can
have an n-dimensional space of different shape and mutually overlap. The architecture of the multlayer adaptive fuzzy probabilistic neural
network, which allow to solve the problem of classification in sequential mode as new data become available, is.proposed. An algorithm for
training the multilayer adaptive fuzzy probabilistic neural network is proposed, and the problem of classification is solved on the basis of the
proposed architecture in terms of intersecting classes, which allows to determine the belonging a single instance of a text document to different
classes with varying degrees of probability. Classifying neural network architecture characterized by simple numerical implementation and high
speed training, and is designed to handle large data sets, characterized by the feature vectors of high dimension. The proposed neural network
and its learning method designed to work in conditions of overlapping classes, differing both the form and size.

References

Specht D. F. Probabilistic neural networks / D. F. Specht // Neural Networks. – 1990. – Vol. 3 (1). – P. 109–118. 2. Бодянский Е. В. Семантическое аннотирование текстовых документов с использованием модифицированной вероятностной нейроной сети / Е. В. Бодянский, О. В. Шубкина // Системные технологии. – Днепропетровск, 2011. – Вып. 4 (75). – С. 48–55. 3. Bodyanskiy Ye. Semantic annotation of text documents using modified probabilistic neural network / Ye. Bodyanskiy, O. Shubkina// Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications: 6th IEEE Intеrnational Conferences, Prague, 15–17 September 2011: – Prague: Czech Technical University In Prague, 2011. – P. 328–331. 4. Bodyanskiy Ye. Semantic annotation of text documents using evolving neural network based on principle «Neurons at Data Points» / Ye. Bodyanskiy, O. Shubkina // Workshop on Inductive Modelling «IWIM 2011» : 4th Interational Conference, Zhukyn-Kyiv, 4–10 July 2011: Kyiv: IRTC ITS, 2011. – P. 31–37. 5. Bodyanskiy Ye. A learning probabilistic neural network with fuzzy inference / Ye. Bodyanskiy, Ye. Gorshkov, V. Kolodyazhniy, J. Wernstedt // Artificial Neural Nets and Genetic Algorithms «ICANNGA 2003» : 6th International Conference, Roanne, France April 23-25 April 2003 : proceedings. – Wien : Springer-Verlag, 2003. – P. 13–17. 6. Bodyanskiy Ye. Resource-allocating probabilistic neuro-fuzzy network / Ye. Bodyanskiy, Ye. Gorshkov, V. Kolodyazhniy // European Union Society for Fuzzy Logic and Technology «EUSFLAT 2003» : 3rd Internetional Conference, Zittau : proceedings. – Zittau : University of Applied Sciences at Zittau/ Goerlitz, 2003. – P. 392-395. 7. Bodyanskiy Ye. Probabilistic neuro-fuzzy network with nonconventional activation functions / Ye. Bodyanskiy, Ye. Gorshkov, V. Kolodyazhniy, J. Wernstedt // Knowledge-Based Intelligent Information and Engineering Systems : 7th International Conference KES 2003, Oxford, 3–5 September 2003 : proceedings. – Berlin-Heidelberg-New York : Springer, 2003. – P. 973–979. – (Lecture Notes in Computer Science, Vol. 2774). 8. Бодянский Е. В. Классификация текстовых документов с помощью нечеткой вероятностной нейронной сети / Е. В. Бодянский, Н. В. Рябова, О. В. Золотухин // Восточно-европейский журнал передовых технологий – 2011. – № 6/2 (54). – С.16–18. 9. Zahirniak D. R. Pattern recognition using radial basis function network / D. R. Zahirniak, R. Chapman, S. K. Rogers, B. W. Suter, M. Kabriski, V. Pyatti // Aerospace Application of Artificial Intelligence: 6 International Conference, 5–8 June 1990 : proceedings. – Dayton : Ohio, 1990. – P. 249–260.

Downloads

Published

2014-12-22

How to Cite

Bodyanskiy, Y. V., Ryabova, N. V., & Zolotukhin, O. V. (2014). MULTILAYER ADAPTIVE FUZZY PROBABILISTIC NEURAL NETWORK IN CLASSIFICATION PROBLEMS OF TEXT DOCUMENTS. Radio Electronics, Computer Science, Control, (1). https://doi.org/10.15588/1607-3274-2015-1-5

Issue

No. 1 (2015): Radio Electronics, Computer Science, Control

Section

Neuroinformatics and intelligent systems

License

Copyright (c) 2015 Ye. V. Bodyanskiy, N. V. Ryabova, O. V. Zolotukhin

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Creative Commons Licensing Notifications in the Copyright Notices

The journal allows the authors to hold the copyright without restrictions and to retain publishing rights without restrictions.

The journal allows readers to read, download, copy, distribute, print, search, or link to the full texts of its articles.

The journal allows to reuse and remixing of its content, in accordance with a Creative Commons license СС BY -SA.

Authors who publish with this journal agree to the following terms:

  • Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License CC BY-SA that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

  • Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

  • Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.