Evaluating Drivers' Smell Preferences with Machine Learning
Author Information
Author(s): Tang Bangbei, Zhu Mingxin, Hu Zhian, Ding Yongfeng, Chen Shengnan, Li Yan
Primary Institution: School of Intelligent Manufacturing Engineering, Chongqing University of Arts and Sciences, Chongqing, China
Hypothesis
Can physiological signals be used to assess drivers' olfactory preferences effectively?
Conclusion
The study shows that a machine learning model can accurately classify drivers' olfactory preferences using physiological signals.
Supporting Evidence
- The decision tree model achieved the highest classification accuracy of 88%.
- Physiological signals were collected in real driving environments to enhance data accuracy.
- Baseline processing improved model accuracy by 3.50% and F1-score by 6.33%.
Takeaway
This study helps understand what smells drivers like while driving, which can make their experience more comfortable.
Methodology
A machine learning model was developed using physiological signals from 33 drivers to classify their olfactory preferences.
Potential Biases
Potential biases due to the limited diversity of the sample and the subjective nature of olfactory preferences.
Limitations
The study's sample size and age range may limit the generalizability of the findings.
Participant Demographics
33 drivers (16 males and 17 females) aged 21 to 32 years with an average driving experience of 3.7 years.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website