Real-Time Data Evaluation with Wearable Devices: an Impact of Artifact Calibration Method on Emotion Recognition

EasyChair Preprint 7313

Abstract

Smartwatch technology is transforming the environment of transmission and monitoring for stakeholders and research participants who want to provide real-time data for evaluation. A range of sensors are available in smartwatches for gathering physical activity and location data. Combining all of these elements allows the collected data to be sent to a remote computer, allowing for real-time physical and perhaps emotional development monitoring. Photoplethysmographic is an uncomplicated and inexpensive optical measurement method that is often used for heart rate monitoring purposes. PPG is a non-invasive technology that uses a light source and a photodetector at the skin's surface to measure the volumetric variations of blood circulation. Models concerning HRV (Heart Rate Variability) analysis are studied in various domains, including human emotion recognition (HER). Smartwatches as sensor-based devices play an essential role as photoplethysmographic (PPG) data are frequently evaluated for this assessment. However, the quality of these signals (in terms of additional disturbances) may not always be optimal, as they are vulnerable to various parameters, such as motion artifacts, light sources, stress distribution, ethnic background, or circumstances. Here techniques for antique rectification play a significant role &, as a response, impact the outcome. This research proposes a novel data distortions mitigation strategy for improving emotion detection classification efficiency using PPG signals during auditory stimulation and a Support Vector Machine (SVM) model. Compared to previously undertaken data using a conventional toolset, i.e.,48.81, the presented scheme improves trigger sensing categorization, i.e., 68.75 percent. An alternative indicator, such as electroencephalographic activity, could be used with PPG for further improvement.

Keyphrases: HRV, PPG, Smartwatch, human emotion support vector machine

@booklet{EasyChair:7313,
  author    = {Fayaz Ahmad Fayaz and Arun Malik},
  title     = {Real-Time Data Evaluation with Wearable Devices: an Impact of Artifact Calibration Method on Emotion Recognition},
  howpublished = {EasyChair Preprint 7313},
  year      = {EasyChair, 2022}}