A systematic hybrid machine learning approach for stress prediction

View article
RT @PeerJCompSci: Ding et al. @EmoryUniversity present a systematic hybrid machine learning approach for stress prediction Read the full…
RT @thePeerJ: Ding et al. @EmoryUniversity present a systematic hybrid machine learning approach for stress prediction Read the full arti…
RT @thePeerJ: Ding et al. @EmoryUniversity present a systematic hybrid machine learning approach for stress prediction Read the full arti…
Ding et al. @EmoryUniversity present a systematic hybrid machine learning approach for stress prediction Read the full article https://t.co/2Njy5jsybE #Bioinformatics #ArtificialIntelligence #DataMining #MachineLearning
RT @PeerJCompSci: Ding et al. @EmoryUniversity present a systematic hybrid machine learning approach for stress prediction Read the full…
Ding et al. @EmoryUniversity present a systematic hybrid machine learning approach for stress prediction Read the full article https://t.co/LroYqHMnqA #Bioinformatics #ArtificialIntelligence #DataMining #MachineLearning
PeerJ Computer Science

Main article text

 

Introduction

  • This study proposed an approach for stress detection which is more accurate and efficient in comparison state-of-the-art approaches.

  • This study proposed a hybrid model that combined two machine learning models using soft voting criteria to achieve significant results in comparison to individual models on the used dataset.

  • Our proposed approach with more several target classes for stress level prediction low/normal, medium-low, medium, medium-high, and high.

  • This study contribute also in terms of efficiency which means the proposed approach gives more accurate results with low computational cost.

  • Present extensive literature on stress detection and give a strong comparison between the state-of-the-art approaches.

  • We have done a statistical T-test to show the significance of the proposed HB model.

Materials and Methods

Dataset description

Data splitting

Machine learning models

LR

RF

GBM

ADA

SVM

HB

Evaluation criteria

Results

Results of machine learning models

Deep learning results

K-fold cross-validation results

Comparison with other studies

Proposed approach results using another dataset

Statistical T-test

  • Accepted null hypothesis means there is no statistical difference in compared results.

  • Accepted alternative hypothesis means there is a statistical difference in compared results.

CONCLUSION

Supplemental Information

Data files for stress detection.

DOI: 10.7717/peerj-cs.1154/supp-1

Code experiments.

DOI: 10.7717/peerj-cs.1154/supp-2

Additional Information and Declarations

Competing Interests

The authors declare that they have no competing interests.

Author Contributions

Cheng Ding conceived and designed the experiments, performed the experiments, analyzed the data, prepared figures and/or tables, authored or reviewed drafts of the article, formatting the document, and approved the final draft.

Yuhao Zhang performed the experiments, analyzed the data, performed the computation work, prepared figures and/or tables, authored or reviewed drafts of the article, and approved the final draft.

Ting Ding conceived and designed the experiments, analyzed the data, performed the computation work, authored or reviewed drafts of the article, and approved the final draft.

Data Availability

The following information was supplied regarding data availability:

The code is available in the Supplemental File.

The data from Kaggle (https://www.kaggle.com/datasets/laavanya/human-stress-detection-in-and-through-sleep, owned by Laavanya Rachakonda from the University of North Carolina Wilmington) is available in the Supplemental File.

Funding

This study was supported by the National Natural Science Foundation of China (Grant No. 41902065). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

17 Citations 2,796 Views 391 Downloads

Your institution may have Open Access funds available for qualifying authors. See if you qualify

Publish for free

Comment on Articles or Preprints and we'll waive your author fee
Learn more

Five new journals in Chemistry

Free to publish • Peer-reviewed • From PeerJ
Find out more