AI SYSTEM FOR DETECTING OPTIMAL CONDITIONS FOR AQUACULTURE

Authors

  • Maria Daskalova
  • Пенка Георгиева

Keywords:

artificial intelligence, machine learning, time series analysis, Black Sea, aquaculture, 1D-CNN.

Abstract

This study presents the development of an artificial intelligence–based system for the analysis and classification of marine environmental. The test dataset includes key physicochemical parameters such as water temperature, salinity, dissolved oxygen, chlorophyll-a concentration, and pH. Data preprocessing involved interpolation of short missing values, removal of longer gaps, and aggregation to daily temporal resolution. A one-dimensional convolutional neural network (1D-CNN) was applied to analyze temporal patterns and perform binary classification of marine conditions as favorable or unfavorable. The results demonstrate that the proposed approach can effectively extract patterns from marine time series and identify periods with favorable environmental conditions.

Downloads

Download data is not yet available.

References

[1] Даскалова М, AI Система за откриване на оптимални условия за аквакултури, Дипломна работа за степен Магистър по Изкуствен интелект, защитена на 11.09.2025г.

[2] Google Developers. Classification: Accuracy, recall, precision, and related metrics, 2024.

https://developers.google.com/machine-learning/crash-course/classification/accuracy-precision-recall

[3] Copernicus Marine MyOcean Viewer . Copernicus.eu. 2024.

https://data.marine.copernicus.eu/viewer/

[4] Yang X, Song Z, Liu J, Zhou C. Deep learning for smart fish farming: applications, opportunities and challenges. ResearchGate . 2020 Apr 6

https://www.researchgate.net/publication/340939333_Deep_learning_for_smart_fish_farming_applications_opportunities_and_challenges

[5] Christenser J, Galeazzi R, Mogensen L, Andersen J. Detection, Localization and Classification of Fish and Fish Species in Poor Conditions using Convolutional Neural Networks.

https://www.researchgate.net/publication/329244872_Detection_Localization_and_Classification_of_Fish_and_Fish_Species_in_Poor_Conditions_using_Convolutional_Neural_Networks. 2018.

[6] Rather MA, Ahmad I, Shah A, Younis Ahmad Hajam, Amin A, Saba Khursheed, Exploring opportunities of Artificial Intelligence in aquaculture to meet increasing food demand. Food chemistry X. 2024 Jun 1;

[7] Verma DK, Singh S, Narendra Kumar Maurya, Kumar P, Rajesh Jayaswal. Important Water Quality Parameters in Aquaculture: An Overview. Agriculture and Environment . 2022 Mar 6;

https://www.researchgate.net/publication/362667844_Important_Water_Quality_Parameters_in_Aquaculture_An_Overview

[8] FAO - Mytilus galloprovincialis . Fao.org. 2024.

https://www.fao.org/fishery/docs/CDrom/aquaculture/I1129m/file/en/en_mediterraneanmussel.htm

[9] Jain A. Understanding the 1D Convolutional Layer in Deep Learning . Medium. 2025.

https://medium.com/@abhishekjainindore24/understanding-the-1d-convolutional-layer-in-deep-learning-7a4cb994c981

[10] xcube Viewer . Brockmann-consult.de. 2025.

https://doors.viewer.brockmann-consult.de/

Published

2026-11-16 — Updated on 2026-10-29

Versions

Issue

Section

Computer Science and Communications - Reviewed Publications. ISSN: 1314-7846

How to Cite

AI SYSTEM FOR DETECTING OPTIMAL CONDITIONS FOR AQUACULTURE. (2026). COMPUTER SCIENCES AND COMMUNICATIONS, 15(1), 55-65. https://ojs.bfu.bg/index.php/CSC/article/view/302 (Original work published 2026)