Using Machine Learning to Model Potential Users with Health Risk Concerns Regarding Microchip Implants.

Understanding traits that are associated with users who are willing to accept microchip implants can help drive future microchip designs, but little is known in this space. We applied three Machine Learning classifiers, that are Decision Trees, Random Forest and XGBoost on demographic information (user characteristics) of 255 survey respondents. The aim was to recognize dominant features and characteristics that lead participants to be categorized as having “Health risk” concern regarding micro-chipping. Comparison of the classifiers in the prediction tasks demonstrated that XGBoost provides the best performance in term of accuracy, precision and recall. XGBoost also showed that “Migration status”, “Race”, “Age” and “Degree” are the most important and “Medical Condition” is the next important characteristic of potential users with “Health risk” concerns about micro-chipping. Further research is needed to classify other concerns and expectations of the survey respondents and to create a fuller understanding of the users willing to accept microchip implants.