User engagement analysis for smart buildings based on social trend tracking

The building sector is one of the major contributors to energy demand. Therefore, policy makers and the public are increasingly aware of the need to address global energy sustainability challenges in the building sector. With the growing prevalence of digitalization and Artificial Intelligence (AI), energy supplying infrastructures are undergoing profound changes by the emergence of smart buildings. Smart buildings aim to not only control devices and appliances inside, but can also create active micro-grids or virtual power plants by adapting to grid's conditions and communicating with other buildings. The great potential to provide flexibility for the integration of fluctuating renewable energy make smart buildings one of the central elements in the future energy system. Although smart building technologies are widely introduced and discussed in academia and industry, its services in real life are still rare. If smart buildings should unfold their potential contribution to the future energy system, one important barrier to overcome is the adoption of these technologies by households and investors. To analyse where the barriers in the adoption decision lie, we apply multiple visual analysis approaches. First, keyword extraction together with network visualization is used to identify the most trending sub-topics of smart building concept. Second, a text mining approach is applied to discover how smart building concepts diffuse over time and spatially based on scraped social media posts and online news. Third, this paper further analyses the user engagement degree in the diffusion of smart building concepts by user profile mining. The paper concludes with deriving recommendations on how (nudging) policies could be designed and how smart building technologies could be enhanced to increase user adaptation.