Scholarlink

<p>The ensemble learning algorithm is a statistical method that is unknown in HEA(s)'s prediction phase. The ensemble learning algorithm is used to check on the phase selection principles, utilizing a large experimental case study on 401 distinct HEAs, comprising 174 SS, 54 IM, and 173 SS + IM phases. Random forest(RF) has the highest accuracy compared with other ensemble learning algorithms i.e. its certainty is about 10% higher than support-vector machines(SVM) and K-nearest neighbors(KNN) for allocating HEA(s). The validity and reliability of the proposed algorithms are announced as results of the paper. Therewith, findings show two main advantages to allocating HEAs: First deduction of decision trees and improving the carefulness, and second automating missing values. In addition, to check the practical accuracy of the machine learning results, the XRD results of the TiZrNbCrV, TiZrNbFeCr, and Ti ZrNbFeV alloys are presented. All alloys are in solid solution statues without any intermetallic phases. The practical results show the ensemble learning algorithms have suitable consistency in real conditions and can be a great help to design new high entropy alloys.</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

The ensemble learning algorithms for prediction high entropy alloys phases

  • 期刊论文

  • 发表日期：

    2025年6月6日

查看全文
  • Masoud Yousefi,
  • Khosrow Rahmani,
  • Masoud Rajabi,
  • Ali Reyhani,
  • Nayereh Asgari

摘要

The ensemble learning algorithm is a statistical method that is unknown in HEA(s)'s prediction phase. The ensemble learning algorithm is used to check on the phase selection principles, utilizing a large experimental case study on 401 distinct HEAs, comprising 174 SS, 54 IM, and 173 SS + IM phases. Random forest(RF) has the highest accuracy compared with other ensemble learning algorithms i.e. its certainty is about 10% higher than support-vector machines(SVM) and K-nearest neighbors(KNN) for allocating HEA(s). The validity and reliability of the proposed algorithms are announced as results of the paper. Therewith, findings show two main advantages to allocating HEAs: First deduction of decision trees and improving the carefulness, and second automating missing values. In addition, to check the practical accuracy of the machine learning results, the XRD results of the TiZrNbCrV, TiZrNbFeCr, and Ti ZrNbFeV alloys are presented. All alloys are in solid solution statues without any intermetallic phases. The practical results show the ensemble learning algorithms have suitable consistency in real conditions and can be a great help to design new high entropy alloys.