Classifying mental workload using EEG data: A machine learning approach


Creative Commons License

Harputlu Aksu Ş., Çakıt E.

13th International Conference on Applied Human Factors and Ergonomics (AHFE 2022) and the Affiliated Conferences, New York, Amerika Birleşik Devletleri, 24 - 28 Temmuz 2022, cilt.42, ss.65-72

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 42
  • Doi Numarası: 10.54941/ahfe1001820
  • Basıldığı Şehir: New York
  • Basıldığı Ülke: Amerika Birleşik Devletleri
  • Sayfa Sayıları: ss.65-72
  • Gazi Üniversitesi Adresli: Evet

Özet

The objectives of this study were two-fold: (1) to investigate the relationship among electroencephalography (EEG) features, task difficulty levels and subjective selfassessment (NASA-Task Load Index (TLX)) scores and (2) to develop machine learning algorithms for classifying mental workload using EEG features. Seventy EEG features (5 frequency band power for 14 channels) were selected as independent variables. One output variable reflecting the difficulty level of n-back memory task was classified. Prefrontal and frontal theta, prefrontal beta-high, occipital, parietal and temporal gamma and occipital alpha activities were found to be the most effective parameters. The results obtained for the four classes of classification problem reached the accuracy of ~68% with Random Forest (RF) algorithm. In addition, maximum accuracy of ~87% was reached with 2-class-based (low and high mental workload) estimation model along with Gradient Boosting Machines (GBM) algorithm. The results from the analysis indicate that EEG signals play an important role in the classification of mental workload. Another remarkable result was high classification performance of GBM, LightGBM and extreme gradient boosting (XGBoost) algorithms that have been developed in the recent past and therefore not frequently used in studies on this subject in the literature.