Scholarlink

The inspection and assessment of water quality hold significant significance in guaranteeing the safety and integrity of our water resources for diverse applications, such as potable consumption, agricultural practices, and industrial utilization. The utilization of machine learning (ML) methods, including transfer learning and the use of transformers, has led to significant progress in the automation of evaluating water quality criteria. This article explores the field of water quality evaluation through the utilization of transfer learning and transformers. It specifically concentrates on significant metrics such as pH value, Hardness, Total Dissolved Solids (TDS), Chloramines, Sulfate, and Organic Carbon. The combination of transfer learning and transformer models may greatly increase the process of water quality evaluation by using the abundant data obtained from Internet of Things (IoT) sensors. Furthermore, we will investigate the potential utilization of well-known transfer learning models, including ResNet-50, DenseNet-121, EfficientNet-B0, Darknet, and VGG16, for this particular objective. Tab-Transformer has a phenomenal accuracy of 94.5%, demonstrating its exceptional capacity to foresee events properly. The excellent capacity of the system is confirmed by the Precision, Recall, and F1 Score values, which are 93.2%, 94.8%, and 94.0%, respectively.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Transfer Learning and Transformer for Multi-parameter Water Quality Assessment: A Novel Framework

  • 章节

  • 发表日期：

    2025年

查看全文
  • Sunil Kumar,
  • Prabal Pratap Singh,
  • Vishal Awasthi,
  • Alok Kumar,
  • Ritesh Agarwal

摘要

The inspection and assessment of water quality hold significant significance in guaranteeing the safety and integrity of our water resources for diverse applications, such as potable consumption, agricultural practices, and industrial utilization. The utilization of machine learning (ML) methods, including transfer learning and the use of transformers, has led to significant progress in the automation of evaluating water quality criteria. This article explores the field of water quality evaluation through the utilization of transfer learning and transformers. It specifically concentrates on significant metrics such as pH value, Hardness, Total Dissolved Solids (TDS), Chloramines, Sulfate, and Organic Carbon. The combination of transfer learning and transformer models may greatly increase the process of water quality evaluation by using the abundant data obtained from Internet of Things (IoT) sensors. Furthermore, we will investigate the potential utilization of well-known transfer learning models, including ResNet-50, DenseNet-121, EfficientNet-B0, Darknet, and VGG16, for this particular objective. Tab-Transformer has a phenomenal accuracy of 94.5%, demonstrating its exceptional capacity to foresee events properly. The excellent capacity of the system is confirmed by the Precision, Recall, and F1 Score values, which are 93.2%, 94.8%, and 94.0%, respectively.