Scholarlink

Data preprocessing deals with those works necessary to put data in machine learning algorithms. Keeping a physics-informed perspective, we also introduce important transformations and filters to improve the performance of learning techniques. Normalization and triggering are discussed, together with convolution filters for moving average and derivatives. The Fourier Transform and the Continuous Wavelet Transform are shown as examples, how a change from the actually data domain helps to uncover features that support machine learning algorithms. This kind of preparations sharpen the view on a problem, help to contrast the right portion of data and provide different views on the data.

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

Data Preprocessing

  • 章节

  • 发表日期：

    2025年

查看全文
  • Marcus J. Neuer

摘要

Data preprocessing deals with those works necessary to put data in machine learning algorithms. Keeping a physics-informed perspective, we also introduce important transformations and filters to improve the performance of learning techniques. Normalization and triggering are discussed, together with convolution filters for moving average and derivatives. The Fourier Transform and the Continuous Wavelet Transform are shown as examples, how a change from the actually data domain helps to uncover features that support machine learning algorithms. This kind of preparations sharpen the view on a problem, help to contrast the right portion of data and provide different views on the data.