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

Publication Date

1-1-2023

Document Type

Conference Proceeding

Publication Title

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume

14051 LNAI

DOI

10.1007/978-3-031-35894-4_42

First Page

574

Last Page

592

Abstract

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.

Keywords

Classification Model, Future user, Health Risk Concern, Implantable, Machine Learning (ML), Prediction, Subcutaneous Microchips, User Characteristics

Department

Mathematics and Statistics

Link to Full Text

Share

COinS