Back to search

Institutional Repository of Vilnius University

Neuroninių tinklų modelių sudarymas Lietuvos vandens telkinių ekologiniams parametrams prognozuoti ir klasifikuoti /

Abstract

dc:description

This thesis investigates the application of machine learning algorithms and artificial neural networks for the prediction and classification of ecological parameters of Lithuanian lakes, based on phytoplankton data. The study focuses on the EFPI (Ecological Phytoplankton Index) and its corresponding ecological class, which plays a crucial role in assessing the ecological status of water bodies. Several machine learning and neural network models were developed in this work: Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), and Long Short­Term Memory (LSTM), as well as classical decision­based models – Decision Tree and Random Forest. These models were evaluated for both EFPI regression and ecological classification tasks using Lithuanian water body data from 2018 to 2023. The experimental results showed that the best EFPI prediction was achieved by the CNN (R2 = 0.624) and LSTM (R2 = 0.586) models, while the highest classification accuracy (0.553) was recorded with the MLP network. It was found that interannual data irregularities had the greatest impact on the LSTM architecture’s performance, while other models remained stable.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Užkurėlis, Marius,

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
lit

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Užkurėlis, Marius,. Neuroninių tinklų modelių sudarymas Lietuvos vandens telkinių ekologiniams parametrams prognozuoti ir klasifikuoti /. Institutional Repository of Vilnius University, 2025. https://repository.vu.lt/VU:ELABAETD239032576&prefLang=en_US