2 results match your criteria: "Beijing Institute of New Technology Applications[Affiliation]"
Missing Value Estimation Methods Research for Arrhythmia Classification Using the Modified Kernel Difference-Weighted KNN Algorithms.
Biomed Res Int
April 2021
School of Mechanical, Electrical and Information Engineering, Shandong University, Weihai, China.
Article Synopsis
- ECG signals are essential for classifying cardiac arrhythmias using machine learning, but datasets often have missing values which complicate classification.
- Multiple methods for estimating these missing values, including Zero, Mean, PCA-based, and RPCA-based methods, are compared in the paper.
- The proposed MKDF-WKNN classification algorithm outperforms existing methods for imbalanced datasets, with RPCA effectively managing missing data in arrhythmia datasets.
Article Synopsis
- The paper proposes a systematic method to evaluate how various traffic conditions affect highway crash risks by using a traffic safety state division approach.
- The study utilizes matched case-control techniques on highway crash and traffic data to minimize external factors' impact on the analysis.
- By employing a multi-parameter fusion cluster method and Bayesian conditional logistic regression, the research identifies distinct traffic safety states and their corresponding crash risk levels, suggesting that focused management can enhance highway safety.