Analysis of Use Cases Enabling AI/ML to IOT Service Platforms

Abstract

Much artificial intelligence (AI) and machine learning (ML) applications use data collected on IoT platforms to train their model. Depending on the quality and quantity of data collected for model training, the performance of AI models varies. The IoT platform is a placeholder for collecting and managing various data such as images, texts, and sensory data among others. Good data management (DM) is very important to building a good AI/ML model. In this paper, we analysed existing AI/ML technologies that can be integrated into an IoT platform. We also investigate potential use cases for AI/ML services that leverage the data collected on the IoT platform. This study analysed existing AI/ML technologies and use cases in a standardized IoT service layer platform, that is oneM2M. Further, in this study potential requirements and key features related to use cases listed by oneM2M have been discussed, which enable AI/ML in the oneM2M system.