A Context-aware Framework for ML Models on Spatio-temporal Data Streams

The rapid development in IoT infrastructure for collecting Spatio-temporal data calls for various machine learning model evolution. Several models can be used on data streams to extract knowledge about different aspects of mobility. To monitor mobility in a dynamic environment like a city or museum, it is necessary to track all ML models deployed in different locations and times. The goal of this research work is to develop a context-aware framework for monitoring mobility in dynamic environments. To develop such a framework, it is essential to define the challenges of managing different ML models on Spatio-temporal data streams.