Skip to content

DidierRLopes/step-detection-ML

Repository files navigation

Step Detection using SVM on NURVV trackers

Notebook Description
Nurvv Classifier Understand Nurvv IMU samples, and if there's a way to classify step/no-step using an unsupervised method.
Nurvv K-Means and Feature Extraction Extract insights from K-Means on Nurvv IMU data and try some feature extraction.
Nurvv from EDA to SVM Reports the entire journey from performing Exploratory Data Analysis (EDA) on the IMU samples, to the signal processing used to label data, and to finally applying Support Vector Machine (SVM) to classify the samples of a model into step or no-step.
Nurvv Pipeline Implement a data pipeline to perform ensemble SVM with multiple models.

Citation

Lopes, D., & Trewartha, G. (2021, December). Step Detection using SVM on NURVV Trackers. In 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA) (pp. 351-356). IEEE.

@inproceedings{lopes2021step,
  title={Step Detection using SVM on NURVV Trackers},
  author={Lopes, Didier and Trewartha, Grant},
  booktitle={2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA)},
  pages={351--356},
  year={2021},
  organization={IEEE}
}