Intelligent Information Bypass for More Efficient Emergency Management

An information bottleneck is one of the typical failure causes in emergency situations. In this paper we focus on a novel approach that consists in applying an intelligent bypass involving natural language processing and AI technologies to minimize the negative effects of the information bottleneck. The system prototype has been implemented for a public security context (anti-hooligan prevention in large scale sport events). It may be adapted to other emergency situations (from anti-terrorist operations to natural disasters). The added value over the standard function of a bypass is the AI-based processing of the information entered to the bypass. As an example of reasoning performed by the bypass we discuss time-and-space information processing implemented in the beta-prototype of the POLINT-112-SMS system. We present the architecture of this system and provide examples of natural language dialogues between human system users and the system (registered at system testing). We discuss a number of possible application areas for the information bypass technology.

---

Opracowanie ze środków MNiSW w ramach umowy 812/P-DUN/2016 na działalność upowszechniającą naukę (zadania 2017).