Метод паралельного формування кортежів ознак даних сегменту складнокласифікованої інформації в умовах невизначеності прийняття рішень та аналізу впливу на розмежованість інформації

МЕТОД ПАРАЛЕЛЬНОГО ФОРМУВАННЯ КОРТЕЖІВ ОЗНАК ДАНИХ СЕГМЕНТУ СКЛАДНОКЛАСИФІКОВАНОЇ ІНФОРМАЦІЇ В УМОВАХ НЕВИЗНАЧЕНОСТІ ПРИЙНЯТТЯ РІШЕНЬ ТА АНАЛІЗУ ВПЛИВУ НА РОЗМЕЖОВАНІСТЬ ІНФОРМАЦІЇ

METHOD OF PARALLEL FORMATION OF TUPLES FEATURES OF DATA IN THE SEGMENT OF DIFFICULT TO CLASSIFY INFORMATION IN CONDITIONS OF UNCERTAINTY MAKING DECISION AND ANALYSIS OF THE IMPACT ON THE INFORMATION FRAGMENTATION

Сторінки: 232-238. Номер: №3, 2021 (297)

Автори:
Е. А. МАНЗЮК
Хмельницький національний університет

E. MANZIUK
Khmelnytskyi National University

DOI:https://www.doi.org/10.31891/2307-5732-2021-297-3-232-238
Надійшла / Paper received :  24.05.2021 р
Надрукована / Paper Printed : 30.06.2021 р

Анотація мовою оригіналу

В роботі наведено результати досліджень з визначення ознак даних, які мають визначальний вплив на розмежованість інформації в умовах невизначеності прийняття рішень інтелектуальними інформаційними системами. Розглядається загальна множина ознак з виділенням підмножини найбільшого впливу на розмежованість даних та формування кортежів ознак в зоні складнокласифікованої інформації при локальному зміщені даних в гіперпросторі ознак.

Ключові слова: інтелектуальна інформаційна технологія, складнокласифікована інформація, прийняття рішень.

Розширена анотація англійською мовою

Detection of atypical data and outliers is an important and difficult task in the classification of information. Information that is considered complex in turn is characterized by a set of features that determine its informativeness. Therefore, the search for signs, due to which the signs of information are considered atypical is an urgent and necessary task. A more difficult task is to find atypical features on a set of limited information, which can be considered data with the availability of alternative solutions, i.e. those data that are located at the class boundaries in the classification of information. Atypical data, zones between classes create difficulties in classifying information and constructing discriminant separation. A method for determining outliers of irrelevant traits based on grouping of class data using the minimum frame tree is proposed. Outliers are detected by minimizing the set of bipartite graph data of adjacent groups. This leads to spatial local demarcation and determination of the set of emission characteristics. The proposed method allows detecting outliers on different sets of features, both common and on the features of individual classes into which the information is divided.

Thus, the tree-like architecture of the computational process of trait research and formation of a set of traits, the change of which leads to compaction of atypical class data allows calculations in parallel and independently with parallelization at the stage of initial initialization class data of different classes.

The method of parallel detection of the set of atypical features of difficult to classify information is implemented. Atypical information is difficult to classify according to classification approaches. However, the analysis of these data is important from the point of view of data detection for studies with variable values of traits. The method allows determining the set of data that significantly affect the delimitation of atypical data.

Keywords: intelligent information technology, difficult to classify information, decision making system.

 

References

1. Zaheer M. Z. Cleaning Label Noise with Clusters for Minimally Supervised Anomaly Detection / M. Z. Zaheer, J. Lee, M. Astrid, A. Mahmood, S.-I. Lee // arXiv:2104.14770 [cs]. — 2021.
2. Sultani W. Real-world Anomaly Detection in Surveillance Videos / W. Sultani, C. Chen, M. Shah // arXiv:1801.04264 [cs]. — 2019.
3. Barmak O. V. Characteristics for choice of models in the ansables classification / O. V. Barmak, Y. V. Krak, E. A. Manziuk // Problems in programming, ISSN 1727-4907. — 2018. — № 2–3. — Pp. 171–179. doi: 10.15407/pp2018.02.171.
4. Krak I. Data Classification Based on the Features Reduction and Piecewise Linear Separation / I. Krak, O. Barmak, E. Manziuk, A. Kulias // Intelligent Computing and Optimization. ICO’2019: Intelligent Computing and Optimization. Advances in Intelligent Systems and Computing. Cham. Intelligent Computing and Optimization. — 2020. — Vol. 1072. — Pp. 282–289. doi: 10.1007/978-3-030-33585-4_28.
5. Manziuk E. A. Approach to creating an ensemble on a hierarchy of clusters using model decisions correlation / E. A. Manziuk, W. Wójcik, O. V. Barmak, I. V. Krak, A. I. Kulias, V. A. Drabovska, V. M. Puhach, S. Sundetov, A. Mussabekova // Przegląd Elektrotechniczny, ISSN 0033-2097. — 2020. — Vol. 96, № 9. — Pp. 108–113. doi: 10.15199/48.2020.09.23.
6. Zvarevashe K. Ensemble Learning of Hybrid Acoustic Features for Speech Emotion Recognition / K. Zvarevashe, O. Olugbara // Algorithms. — 2020. — Vol. 13, № 3. — Pp. 70. doi: 10.3390/a13030070.
7. Alexandropoulos S.-A. N. A new ensemble method for outlier identification / S.-A. N. Alexandropoulos, S. B. Kotsiantis, V. E. Piperigou, M. N. Vrahatis // 2020 10th International Conference on Cloud Computing, Data Science Engineering (Confluence). — 2020. — Pp. 769–774. doi: 10.1109/Confluence47617.2020.9058219.
8. Krak I. Approach to Piecewise-Linear Classification in a Multi-dimensional Space of Features Based on Plane Visualization / I. Krak, O. Barmak, E. Manziuk, H. Kudin // Lecture Notes in Computational Intelligence and Decision Making. ISDMCI’2019: Advances in Intelligent Systems and Computing. Advances in Intelligent Systems and Computing. Cham. Lecture Notes in Computational Intelligence and Decision Making. — 2020. — Vol. 1020. — Pp. 35–47. doi: 10.1007/978-3-030-26474-1_3.
9. Oosterlinck D. From one-class to two-class classification by incorporating expert knowledge: Novelty detection in human behaviour / D. Oosterlinck, D. F. Benoit, P. Baecke // European Journal of Operational Research. — 2020. — Vol. 282, № 3. — Pp. 1011–1024. doi: 10.1016/j.ejor.2019.10.015.
10. Abati D. Latent Space Autoregression for Novelty Detection / D. Abati, A. Porrello, S. Calderara, R. Cucchiara // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019. — Pp. 481–490.
11. Rodríguez-Ruiz J. A one-class classification approach for bot detection on Twitter / J. Rodríguez-Ruiz, J. I. Mata-Sánchez, R. Monroy, O. Loyola-González, A. López-Cuevas // Computers & Security. — 2020. — Vol. 91. — Pp. 101715. doi: 10.1016/j.cose.2020.101715.
12. Gong D. Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection / D. Gong, L. Liu, V. Le, B. Saha, M. R. Mansour, S. Venkatesh, A. van den Hengel // Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019. — Pp. 1705–1714.
13. You C. Provable Self-Representation Based Outlier Detection in a Union of Subspaces / C. You, D. P. Robinson, R. Vidal // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017. — Pp. 3395–3404.
14. Lai C.-H. Robust Subspace Recovery Layer for Unsupervised Anomaly Detection / C.-H. Lai, D. Zou, G. Lerman // arXiv:1904.00152 [cs, stat]. — 2019.
15. Barmak O. Using piecewise hyper linear classification in multidimensional feature space for text content / O. Barmak, E. Manziuk, I. Krak // 2019 IEEE 14th International Conference on Computer Sciences and Information Technologies (CSIT). — 2019. — Vol. 2. — Pp. 119–123. doi: 10.1109/STC-CSIT.2019.8929798.
16. Barmak O. Classification Based Hierarchical Clustering Prediction Variability in the Ensembles of Models Using a Statistical Approach / O. Barmak, E. Manziuk, I. Krak // 2020 IEEE 15th International Conference on Computer Sciences and Information Technologies (CSIT). — 2020. — Vol. 1. — Pp. 11–14. doi: 10.1109/CSIT49958.2020.9322019.
17. Barmak O. ‪Diversity as The Basis for Effective Clustering-Based Classification / O. Barmak, I. Krak, E. Manziuk // CEUR Workshop Proceedings. ICST’2020: 9th International Conference “Information Control Systems & Technologies. Odessa, Ukraine. CEUR Workshop Proceedings. — 2020. — Vol. 2711. — Pp. 53–67.
18. Cheng Z. Unsupervised Outlier Detection via Transformation Invariant Autoencoder / Z. Cheng, E. Zhu, S. Wang, P. Zhang, W. Li // IEEE Access. — 2021. — Vol. 9. — Pp. 43991–44002. doi: 10.1109/ACCESS.2021.3065838.