Developing Test Scenarios and Test Oracle for AutoML Applications on the Intelligent Edge

Sammanfattning

Artificial Intelligence (AI) and Machine Learning (ML) are the most popular discussed subject nowadays which in turn are the core of this project. This project is developing the test cases that can be applied for machine learning applications by using different machine learning models and Automated Machine Learning (AutoML). In addition, AutoML is a cutting-edge technology designed to streamline the various tasks and processes associated with machine learning. This technology provides computer scientists with an exceptional opportunity to develop machine learning models that are both more efficient and of higher quality. The main idea which this project tries to handle is developing test scenarios for AutoML applications that the industry uses for their Heating, Ventilation, and Air Conditioning (HVAC) system. This project addresses the industry’s request for research that can determine which machine learning model is best suited for the HVAC system in Intelligent Edge. The company began with a focus on indoor air quality. Initially focusing on air quality monitors, they pivoted in 2021 to develop sensors that also analyze occupancy and control building systems. Their mission is to create a balanced workplace that fosters mutual respect, empowerment, and autonomy for the team. It handles real data and conditions from the industry system to predict the temperature for the next hour. The goal of this project is to identify a machine learning model that can predict the next hour’s temperature. The project identifies the test scenarios for a smart building, incorporating factors such as weather conditions, time variations, and energy usage. These comprehensive scenarios are then tested via various machine learning models, including AutoML, XGBoost, and Linear regression, using real-time data to predict the temperature for the next hour.