Background

Social Media has become a primary source of information, with citizens sharing information on ongoing situations.

Technology Overview

The FÍOSÍN Social Media Monitor automatically aggregates and filters relevant messages into actionable information. Control room technologies currently monitor various streams of information (phones, webcams, GIS maps), but they have not yet effectively integrated social media data. Featuring advanced methods in machine learning as well as natural language processing, our platform filters noise to assist decision making.

FÍOSÍN uses machine learning techniques to automatically improve its own algorithm while data is being collected ( Figure 1. ) ( Figure 2. ).

The system assists in:

distribution of resources
aids communication strategy
verifying existing information
generating data for reporting and training

When Hurricane Matthew made landfall in 2016, the researchers recorded 1.2 million data points (text and image) from Twitter, and filtered to 6,000 relevant messages over a 6-day period.

Stage of Development:

Ready for market. Pilot deployments are live with emergency services in Ireland and Northern Italy.

Benefits

FÍOSÍN filters noise to only keep relevant messages
Messages are displayed clearly on a map to assist decision making
FÍOSÍN harnesses data from a range of sources and can analyse both image and text.
The system monitor automatically highlights key messages and removes irrelevant or untrustworthy information.

Applications

When an event (emergency or planned major event) is taking place, any target geographical region can be selected for real-time monitoring.

The FÍOSÍN Social Media Monitor has been tested with users in:

Emergency Management: response and recovery
Business Continuity: planning and preparedness
Reputation management: communication and mitigation

FÍOSÍN: Social Media Analysis for Decision Making in Emergency Management

IP Status

Seeking