Explainable quantum inspired LSTM model for water quality classification in fishponds
Abstract
Water quality monitoring is a crucial aspect of aquaculture, contributing significantly to the health, productivity, and sustainable management of aquatic environments. It is important to monitor key parameters like pH, dissolved oxygen, turbidity and temperature levels continuously to ensure optimal water conditions. The traditional monitoring methods, however, tend to be costly, time-consuming, and sometimes ineffective when trying to monitor a dynamic aquatic system in real-time. Recently, several deep learning and quantum inspired algorithms such as ANN, CNN, LSTM, GRU, QSVM and QDEDL-WQI frameworks have been developed for water quality prediction and classification. These approaches, although showing good performance, have still not overcome issues of nonlinearity in the environment, learning long-term temporal dependencies, and computational efficiency. For this reason, this paper presents an intelligent water quality classification system based on the Quantum-inspired LSTM (QLSTM) for aquaculture systems. It is suggested that the proposed QLSTM model include quantum superposition-based feature encoding and quantum entanglement mechanisms to augment feature representation and learning ability. The proposed model attained an accuracy of 99.63%, recall of 98.66%, precision of 99.32%, an F1-score of 98.99%, and a loss of 0.0291. Proposed QLSTM is better than the existing deep learning and quantum-enhanced models. The framework provides for efficient and reliable real-time water quality monitoring and sustainable aquaculture management.
Related
- A Taxonomy-Inspired Hierarchical Classification of Plastic Litter
- Advancing Near-Term Water Quality Forecasting with Explainable Machine Learning and Probabilistic Uncertainty Quantification
- Explainable AI with dual-layer validation framework for ensuring noise resilience and physical consistency in low-cost sustainable water quality monitoring
- Machine learning models with explainable AI for PM2.5 prediction for Urban air quality assessment
- An explainable swarm-optimized hybrid GRU–Transformer framework with wavelet-based spatial–temporal feature modeling and activation function analysis for rainfall classification across Pakistan’s climatic regimes
BibTeX
@article{arepalli2026explainable,
title = {{Explainable quantum inspired LSTM model for water quality classification in fishponds}},
author = {Peda Gopi Arepalli and Chin-Shiuh Shieh and Satyanarayana Sanakkayala and Kiran Siripuri},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-71213-x},
url = {https://www.nature.com/articles/s41598-026-71213-x},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Explainable quantum inspired LSTM model for water quality classification in fishponds
AU - Peda Gopi Arepalli
AU - Chin-Shiuh Shieh
AU - Satyanarayana Sanakkayala
AU - Kiran Siripuri
PY - 2026
DA - 2026/10/05
JO - Scientific Reports
DO - 10.1038/s41598-026-71213-x
UR - https://www.nature.com/articles/s41598-026-71213-x
AB - Water quality monitoring is a crucial aspect of aquaculture, contributing significantly to the health, productivity, and sustainable management of aquatic environments. It is important to monitor key parameters like pH, dissolved oxygen, turbidity and temperature levels continuously to ensure optimal water conditions. The traditional monitoring methods, however, tend to be costly, time-consuming, and sometimes ineffective when trying to monitor a dynamic aquatic system in real-time. Recently, several deep learning and quantum inspired algorithms such as ANN, CNN, LSTM, GRU, QSVM and QDEDL-WQI frameworks have been developed for water quality prediction and classification. These approaches, although showing good performance, have still not overcome issues of nonlinearity in the environment, learning long-term temporal dependencies, and computational efficiency. For this reason, this paper presents an intelligent water quality classification system based on the Quantum-inspired LSTM (QLSTM) for aquaculture systems. It is suggested that the proposed QLSTM model include quantum superposition-based feature encoding and quantum entanglement mechanisms to augment feature representation and learning ability. The proposed model attained an accuracy of 99.63%, recall of 98.66%, precision of 99.32%, an F1-score of 98.99%, and a loss of 0.0291. Proposed QLSTM is better than the existing deep learning and quantum-enhanced models. The framework provides for efficient and reliable real-time water quality monitoring and sustainable aquaculture management.
SN - 2045-2322
ER -