AI-powered operating system for buildings: new performance metrics

In the real estate sector, a large volume of data is produced by businesses, commercial users and building visitors in a great variety of forms. For instance, three extensive sources of data come from unstructured text (e.g. documents, contracts), numerical data containing resources consumption and sensor/image-type data describing user behavior. A challenging problem for the sector is how to process the generated data into a useful asset that can provide insights to help business decisions, optimize user navigation and automate building-related processes. In this context, modern machine learning offers a plethora of solutions, ranging from well-established predictive statistical models to natural language processing techniques for automatically analyzing texts. Given the variety of out-of-the box machine learning methods, it becomes essential to be able to compare solutions efficiently in terms of quality measure, model training and generalization capacity for annotated data, as well as intrinsic metrics for non-annotated data. TO BE CONT'D

Intern: 
Yan Fu
Faculty Supervisor: 
Ashish Khisti
Province: 
Ontario
Partner: 
Partner University: 
Program: