Siamese Network-Based Transfer Learning Model to Predict Geogenic Contaminated Groundwaters

Abstract

Exposure to geogenic contaminated groundwaters (GCGs) is a significant public health concern. Machine learning models are powerful tools for the discovery of potential GCGs. However, the insufficient groundwater quality data and the fact that GCGs are typically a minority class in data hinder models to produce meaningful GCG predictions. To address this issue, a deep learning method, Siamese network-based transfer learning (SNTL), is used to estimate the probability that hazardous substances are present in groundwater above a threshold based on limited and class-imbalanced data. SNTL greatly reduces the amount of required training data and eliminates negative effects of class-imbalanced data on prediction model performance. The predictions of three typical GCGs (high arsenic/fluoride/iodine groundwater) show that the SNTL models provide higher (about 80%) and more balanced sensitivity and specificity than benchmark Random Forest models, indicating that SNTL models can predict both GCGs and non-GCGs. Therefore, protecting populations from GCG exposure in areas where other prediction methods fail to contribute risk information due to poor groundwater quality data can be enabled by SNTL.

Prediction of Activated Sludge Sedimentation Performance Using Deep Transfer Learning

“…This study also identified some key limitations of the proposed method. The experimental results generally showed that the recognition performance of models for the nonbulking sludge category was significantly higher than for the bulking category, mainly due to the difference in sample data volume in different categories . In addition, the collected sludge samples only represent their morphological characteristics at the time of sampling without providing any information about long-term variation in characteristics.…”

Integrating sensor data and machine learning to advance the science and management of river carbon emissions

“…A second issue concerns detection, as the accuracy of DL methods relies on the quantity of observational data. Insufficient data may prevent DL from achieving satisfactory precision (Cao et al, 2022); however, even in developed countries with well-established infrastructures, the cost of obtaining a substantial volume of highprecision environmental monitoring data such as that needed for river carbon cycle estimation could hinder the application of DL in some locations (Richards et al, 2023). Moreover, even water quality monitoring networks in developing countries are often limited by financial resources and technical capabilities and so must prioritize resource allocation.…”

Integrated Transfer Learning and Multitask Learning Strategies to Construct Graph Neural Network Models for Predicting Bioaccumulation Parameters of Chemicals

“…11−13 However, TL was primarily applied to a single task and often overlooks potential relationships that may exist between different end points. 11 Another ML paradigm, MTL, can overcome these limitations. MTL leverages experience from related tasks to facilitate the learning of specific tasks, thereby enhancing prediction accuracy by learning multiple related tasks jointly and reducing the risks of overfitting.…”

“…14−16 TL and MTL have been successfully applied separately to construct models for environmentally related end points and were proved to improve model performance. 11,17 It can be inferred that their joint application can further improve the model performance, especially in cases with limited data.…”

“…TL is a ML paradigm that employs a pretrained model, typically trained on large data sets, as a starting point to train new models for specific target tasks. Instead of starting from scratch, TL allows to utilize knowledge and insights gained from a different but related task, thereby improving the performance of the target tasks. − However, TL was primarily applied to a single task and often overlooks potential relationships that may exist between different end points . Another ML paradigm, MTL, can overcome these limitations.…”