Scholarlink

<p>Water is a most important resource and essential to sustain creatures. The undesirable effect of water pollution is a major environmental concern on both life and human beings. Water pollution influences several factors such as water quality; human health, economic development, and social wealth. Therefore, the identification of the groundwater quality is necessary to preserve the water resources for various applications. But, the existing methods failed to improve the accurate quality prediction with lesser time consumption. Therefore, a novel technique called Generalized Reciprocal Groundwater Identification based Tversky Indexive Support Vector eXtreme Gradient Boost Classification (GRGI-TISVXGBC) is introduced for performing the quality prediction analytics by better accuracy as well as minimal time. Groundwater identification and quality prediction analytics are two process of GRGI-TISVXGBC. The Generalized Reciprocal Method (GRM) is used in GRGI-TISVXGBC technique for identifying the groundwater. After identifying the presence of groundwater, the quality analysis is carried out by using Tversky Indexive Support Vector eXtreme Gradient boosting (TISXGB) classification method. Support Vector eXtreme Gradient Boosting Classification is carried out to form the strong classifier through combining the weak learner for water quality analysis. In the designed technique, SVM was considered to analyze the data parameters like temperature, pH, nitrates for groundwater quality prediction analysis using Tversky similarity Index. Weak learners were gathered to form strong classifier by better accuracy as well as lesser error rate using gradient descent step-size function. It aids to improve groundwater quality prediction. Experimental evaluation is conducted on parameters namely prediction accuracy, false-positive rate, as well as time using number of data. Results analysis confirms the GRGI-TISVXGBC enhances quality prediction accuracy as well as lesser false-positive rate, prediction time as compared to conventional techniques.</p>

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

Generalized Reciprocal Based Tversky Indexive Support Vector Extreme Boost Classification for Water Quality Prediction Analysis

  • 期刊论文

  • 发表日期：

    2026年2月25日

查看全文
  • Raghuveer Narsing,
  • Saikumar Chary Konnoju

摘要

Water is a most important resource and essential to sustain creatures. The undesirable effect of water pollution is a major environmental concern on both life and human beings. Water pollution influences several factors such as water quality; human health, economic development, and social wealth. Therefore, the identification of the groundwater quality is necessary to preserve the water resources for various applications. But, the existing methods failed to improve the accurate quality prediction with lesser time consumption. Therefore, a novel technique called Generalized Reciprocal Groundwater Identification based Tversky Indexive Support Vector eXtreme Gradient Boost Classification (GRGI-TISVXGBC) is introduced for performing the quality prediction analytics by better accuracy as well as minimal time. Groundwater identification and quality prediction analytics are two process of GRGI-TISVXGBC. The Generalized Reciprocal Method (GRM) is used in GRGI-TISVXGBC technique for identifying the groundwater. After identifying the presence of groundwater, the quality analysis is carried out by using Tversky Indexive Support Vector eXtreme Gradient boosting (TISXGB) classification method. Support Vector eXtreme Gradient Boosting Classification is carried out to form the strong classifier through combining the weak learner for water quality analysis. In the designed technique, SVM was considered to analyze the data parameters like temperature, pH, nitrates for groundwater quality prediction analysis using Tversky similarity Index. Weak learners were gathered to form strong classifier by better accuracy as well as lesser error rate using gradient descent step-size function. It aids to improve groundwater quality prediction. Experimental evaluation is conducted on parameters namely prediction accuracy, false-positive rate, as well as time using number of data. Results analysis confirms the GRGI-TISVXGBC enhances quality prediction accuracy as well as lesser false-positive rate, prediction time as compared to conventional techniques.