Multipopulation Whale Optimization-Based Feature Selection Algorithm and Its Application in Human Fall Detection Using Inertial Measurement Unit Sensors
2024

Whale Optimization Algorithm for Feature Selection in Human Fall Detection

Sample size: 157 publication 10 minutes Evidence: high

Author Information

Author(s): Cao Haolin, Yan Bingshuo, Dong Lin, Yuan Xianfeng

Primary Institution: Shandong University, Weihai, China

Hypothesis

Can a multispiral whale optimization algorithm improve feature selection for human fall detection?

Conclusion

The proposed multispiral whale optimization algorithm outperforms existing methods in feature selection for human fall detection.

Supporting Evidence

  • The MSWOA showed superior performance in feature selection compared to six other algorithms.
  • MSWOA achieved a higher classification accuracy on 17 out of 20 datasets.
  • The algorithm effectively reduced the number of selected features while maintaining accuracy.
  • MSWOA was tested on a dataset specifically designed for human fall detection.

Takeaway

This study created a new algorithm to help computers pick the best features for detecting when someone falls, making it better at recognizing falls.

Methodology

The study used a multispiral whale optimization algorithm tested on 20 UCI datasets and a human fall detection dataset.

Potential Biases

Potential bias in dataset selection and participant demographics.

Limitations

The algorithm may not perform as well on all datasets, particularly those with complex feature distributions.

Participant Demographics

Seven participants, including two females and five males, average age 22 years.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.3390/s24247879

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication