A Neutrosophic Triplet-Plithogenic Model for Evaluating Talent Training Quality in IoT Application Technology Based on Artificial Intelligence

Abstract

Evaluating the quality of Internet of Things (IoT) talent training in vocational 
colleges is critical yet challenging due to the diverse, interdependent skills required and 
the complexities of blended learning environments. This study proposes a novel 
neutrosophic triplet-plithogenic model to assess training effectiveness, integrating 
neutrosophic triplet sets to capture truth, indeterminacy, and falsehood in student 
performance and plithogenic sets to model skill interdependencies via contradiction 
degrees. The model aggregates evaluations from instructors, AI systems, and students, 
adjusted for reliability, to generate granular proficiency profiles and class-level quality 
indices. Applied to scenarios like network security, smart agriculture, and industrial IoT, 
and culminating in a comprehensive smart city training program evaluation, the model 
provides actionable insights for curriculum optimization. Results demonstrate its ability 
to identify skill gaps, address uncertainties, and align training with industry needs. 
Limitations, such as parameter sensitivity, and future directions, including empirical 
validation, are discussed, positioning the model as a transformative tool for IoT education. 

DOI: 10.5281/zenodo.15635468