Many characteristics of Smart cities rely on a sufficient quantity and quality of urban data. Local industry and developers can use this data for application development that improves life of all citizens. Therefore, the handling and usability of this data is a big challenge for smart cities. In this paper we investigate new approaches to urban data management using emerging technologies and give an insight on further research conducted within the EC-funded smarticipate project. Geospatial data cannot be handled well in classical relational database environments. Either they are just put in as binary large objects or have to be broken down into elementary types which can be handled by the database, in many cases resulting in a slow system, since the database technology is not really tuned for delivery on mass data as classical relational databases are optimized for online transaction processing and not analytic processing.
In this paper we analyse key features of open geodata portals such as linked data and machine learning in order to show ways of improving the user experience. Based on the Smarticipate projects we show afterwards as open data and geo data online and see the practical application. We also give an outlook on piloting cases where we want to evaluate, how the technologies presented in this paper can be combined to a usefull open data portal. In contrast to the previous EC-funded project urbanapi, where participative processes in smart cities where created with urban data, we go one step further with semantic web and open data. Thereby we achieve a more general approach on open data portals for spatial data and how to improve their usability.