*Risk Assessment and Automated Anomaly Detection Using a Deep Learning Architecture DOI: http://dx.doi.org/10.5772/intechopen.96209*

services (for products, point of interests, etc.), notifications and alerts. Finally, via these devices each user may provide a feedback to the system about requests or reports, about incidents that may concern their safety, or public security, or interactions with the system in the context of system automatic personal servicing.

The back–end component contains the intelligence modules which process the input coming from sensing devices and produce high level intelligence and metadata which assist operational personnel, enhance end–users' experience and content management services. These metadata are used either for further processing by fusion algorithms, or presented to end – devices via visualization methods on each end–device. Such metadata concerns directed paths for navigation services, fused high level visual information, or information regarding recommendations, detected incidents or notifications and alerts. Finally, the content management services enable efficient data storing and retrieving operations in a scalable way. The back – end component comprises a Message-oriented middleware in order to interconnect all the sensing and processing component, provides a REST API to front–end devices, supports web platforms interfaces (web – portal) and orchestrates the accurate functionality of the whole system, **Figure 23**.

The core intelligence residing in the "Analytics, Data Fusion and Risk-based Security Server" is presented in Section 3.3. The Data protection, Legal Compliance and Ethics are important aspects that should be taken into consideration in the system architecting process and are analyzed in Section 3.4.

## *3.2.2 Analytics, data fusion and risk-based security server*
