APPLICATION OF DECISION TREE ALGORITHMS IN THE PROBLEMS OF INTERPRETED CLASSIFICATION OF HYDROACOUSTIC SIGNALS

Abstract

The article considers the application of decision tree algorithms for the tasks of interpreted classification of
hydroacoustic signals in complex marine environment conditions. The relevance of the study is due to the need to combine
automated analysis of underwater acoustic data with the ability to explain the classification results, which is critically
important for monitoring and security systems. The main attention is paid to the rule-based approach, which allows
forming classification solutions based on logical rules with a clear physical and semantic interpretation, without using
complex neural models of the "black box" type. The paper analyzes the features of hydroacoustic signals as an object of
classification, in particular their non-stationary nature, noise and dependence on the parameters of the aquatic
environment. A generalized scheme for constructing a decision tree based on acoustic features of signals, such as spectral
characteristics, intensity level, statistical parameters and temporal features, is proposed. It is shown that each path in the
decision tree can be represented as a rule-based rule of the “if–then” type, which ensures transparency and controllability
of the decision-making process. An example of practical application of a combination of statistical analysis, threshold
rule-based classification and machine learning methods for processing hydroacoustic data of the northwestern part of the
Black Sea is considered. It is shown that the use of decision trees allows you to work effectively under conditions of
limited training samples, reduce the risk of overtraining and adapt the model to expert rules. The results obtained confirm
the feasibility of using decision tree algorithms in practical hydroacoustic monitoring systems, where the key requirements
are interpretability, adaptability and reliability of classification.
Keywords: decision trees, hydroacoustic signals, classification, rule-based approach, interpreted algorithms,
machine learning, software.

References
1. Верлань А. І., Олексій А. О. Огляд та порівняння методів машинного навчання для розпізнавання
гідроакустичних сигналів. Інфокомунікаційні та комп’ютерні технології. 2022. Т. 1, № 03. С. 296–306. DOI:
https://doi.org/10.36994/2788-5518-2022-01-03-18. URL: https: // visn-icct.uu. edu. ua/ index.php / icct / article/view/83.
Дата звернення: 14.05.2026.
2. Xie Y., Ren J., Xu J. Guiding the underwater acoustic target recognition with interpretable contrastive learning.
OCEANS 2023 – Limerick. 2023. DOI: https://doi.org/10.1109/OCEANSLimerick52467.2023.10244447. Дата
звернення: 14.05.2026.
3. Wang M., Zhu Z., Qian G. Modulation Signal Recognition of Underwater Acoustic Communication Based on
Archimedes Optimization Algorithm and Random Forest. Sensors. 2023. Vol. 23, No. 5. Article 2764. DOI: https://doi.
org/10.3390/s23052764. Дата звернення: 14.05.2026.
4. Luo X., Chen L., Zhou H., Cao H. A Survey of Underwater Acoustic Target Recognition Methods Based on
Machine Learning. Journal of Marine Science and Engineering. 2023. Vol. 11, No. 2. Article 384. DOI: https://
doi.org/10.3390/jmse11020384. Дата звернення: 14.05.2026.
5. Schwardt M., Pilger C., Gaebler P., Hupe P., Ceranna L. Natural and Anthropogenic Sources of Seismic,
Hydroacoustic, and Infrasonic Waves: Waveforms and Spectral Characteristics and Their Applicability for Sensor
Calibration. Surveys in Geophysics. 2022. Vol. 43, No. 5. P. 1265–1361. DOI: https://doi.org/10.1007/s10712-022-
09713-4. Дата звернення: 14.05.2026.
6. Song Y.-Y., Lu Y. Decision tree methods: applications for classification and prediction. Shanghai Archives of
Psychiatry. 2015. Vol. 27, No. 2. P. 130–135. DOI: https://doi.org/10.11919/j.issn.1002-0829.215044. URL:
https://pmc.ncbi.nlm.nih.gov/articles/PMC4466856/. Дата звернення: 14.05.2026.
7. Lee S., Comuzzi M., Kwon N. Exploring the Suitability of Rule-Based Classification to Provide Interpretability
in Outcome-Based Process Predictive Monitoring. Algorithms. 2022. Vol. 15, No. 6. Article 187. DOI:
https://doi.org/10.3390/a15060187. URL: https://www.mdpi.com/1999-4893/15/6/187. Дата звернення: 14.05.2026.
8. Geng X., Yang Z., Jiao L., Zhou Z.-J., Ma Z. Association Rule-Based Classification: A Comprehensive Review
of Methodologies and Applications. Expert Systems with Applications. 2025. Vol. 280. Article 127454. DOI:
https://doi.org/10.1016/j.eswa.2025.127454. URL: https:// www.sciencedirect.com /science / article/ abs/pii/S09574174-
25010760. Дата звернення: 14.05.2026.
9. Tran K. Human-Learn: Rule-Based Learning as an Alternative to Machine Learning. Medium. 2023. URL:
https: // medium.com / data-science / human-learn-rule-based-learning-as-an-alternative-to-machine-learning-baf1899-
ecb3a. Дата звернення: 14.05.2026.
10. Montgomery R. M. A Comparative Analysis of Decision Trees, Neural Networks, and Bayesian Networks:
Methodological Insights and Practical Applications in Machine Learning. Preprints.org. 2024. DOI:
https://doi.org/10.20944/preprints202410.1491.v1. URL: https://www.preprints.org/manuscript/202410.1491. Дата
звернення: 14.05.2026.
11. Mihailov M. E. Characterization and Automated Classification of Underwater Acoustic Environments in the
Western Black Sea Using Machine Learning Techniques. Journal of Marine Science and Engineering. 2025. Vol. 13, No.
7. Article 1352. DOI: https://doi.org/10.3390/jmse13071352. Дата звернення: 14.05.2026.