Scholarlink

<p>We introduce Automata Pattern (AutPat), a feature extractor for EEG, and embed it in an explainable feature engineering (XFE) pipeline.&#xa0;We evaluated AutPat on three tasks: EEG artifact classification, stress detection, and mental performance detection. The pipeline computes AutPat features from raw EEG, selects informative variables with cumulative weighted iterative neighborhood component analysis (CWINCA), and performs classification using a t-algorithm-based k-nearest neighbors (tkNN) classifier. For interpretability, we map the selected features to Directed Lobish (DLob) symbols and derive DLob strings and cortical connectome diagrams.&#xa0;The AutPat-based XFE achieved &gt; 88% classification accuracy on all datasets. CWINCA reduced the feature space while maintaining accuracy, and the DLob layer yielded dataset-specific symbolic outputs and 8 × 8 connectome matrices.&#xa0;AutPat, combined with CWINCA and tkNN, provides a compact and accurate EEG pipeline with inherent symbolic explanations. The results indicate that AutPat-based XFE is a practical option for EEG analysis when both performance and interpretability are required.</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

An Explainable Feature Engineering Model Based on Automata Pattern: Investigations on the EEG Artifact Classification

  • 期刊论文

  • 发表日期：

    2025年11月4日

查看全文
  • Irem Tasci,
  • Sengul Dogan,
  • Turker Tuncer

摘要

We introduce Automata Pattern (AutPat), a feature extractor for EEG, and embed it in an explainable feature engineering (XFE) pipeline. We evaluated AutPat on three tasks: EEG artifact classification, stress detection, and mental performance detection. The pipeline computes AutPat features from raw EEG, selects informative variables with cumulative weighted iterative neighborhood component analysis (CWINCA), and performs classification using a t-algorithm-based k-nearest neighbors (tkNN) classifier. For interpretability, we map the selected features to Directed Lobish (DLob) symbols and derive DLob strings and cortical connectome diagrams. The AutPat-based XFE achieved > 88% classification accuracy on all datasets. CWINCA reduced the feature space while maintaining accuracy, and the DLob layer yielded dataset-specific symbolic outputs and 8 × 8 connectome matrices. AutPat, combined with CWINCA and tkNN, provides a compact and accurate EEG pipeline with inherent symbolic explanations. The results indicate that AutPat-based XFE is a practical option for EEG analysis when both performance and interpretability are required.