CityPulse is a research project on innovative smart city applications to help people and organisations make more informed decisions in their day-to-day tasks involving public services by providing tools to enrich, integrate and process (near-) real-time data from smart cities. It is also implementing use-case and demonstrators in the partner cities (Aarhus and Brasov) for smart journey, social media event extraction, city dashboard and smart transport to show-case the results of the project.
The City Pulse demo at the Mobile World Congress in Barcelona (European Commission pavilion - Hall 4 at the Congress Square) will present a mobile app that provides online information about events and analysis of data in a smart city. This allows users to find a route from a source to a destination in the city of Aarhus in Denmark. Users can also choose the types of situations they want to see or to avoid (e.g. air quality related situations, pollen levels, traffic congestion) and they can also find a suitable parking spot based on live information collected by sensory devices. A set of back-end services is responsible for collecting and integrating various sensory data from the city environment and analytical components, converting them to higher-level abstractions and actionable-information that are shown on the mobile app.
The second part of the demo includes a data quality analysis, a 3D map and data visualization tool, and a deep learning software for social media analysis that uses relevant Twitter data (e.g. citizen sensory information such at traffic incidents, social events, and crime), and extracts various events or conditions reported by people on social media. The social media analysis software will show a live demo of events extracted from the city of London from Twitter streams.
In the future, City Pulse technologies could help enrich data streams from physical sensing devices with metadata annotations and enable adaptive processing, aggregation and federation of data. Functionalities for aggregation and federation using linked data and mash-up techniques enable the development of a scalable framework for processing large-scale Internet of Things data streams in smart city environments.