Abstract
In today's busy world, users and authorities require better services to achieve their daily activities and tasks in a smart way by using available resources in an optimized manner. The variety of available data sources, starting from crowdsourced data, open governmental data, and other online sources can provide users with smart tools to better manage their daily activities. However, collecting and integrating this multitude of overlapping data sources is a challenging task. Particularly, digital maps are being extensively used to browse and share information about points of interest, plan trips, and to find optimized paths. Within this context, there is a real opportunity to enrich traditional maps with different knowledge-based layers extracted from the variety of available data sources. This paper introduces the concept of "smart maps" by collecting, managing, and integrating heterogeneous data sources in order to infer relevant knowledge-based layers. Unlike conventional maps, smart maps extract live events (e.g., concert, competition, incident), online offers, and statistical analysis (e.g., dangerous areas) by encapsulating incoming semi-and unstructured data into structured generic packets. These packets are processed to extract statistical knowledge on accident-prone and safe areas, and detect Events of Interest (EoI), based on a multi-dimensional clustering technique. This approach lays the ground for delivering different intelligent services and applications, such as: 1) city explorer that provides latest information collected from multiple sources about places and events; and 2) route and trip planning that leverage smart map framework to recommend safe routes.