Using event representation and semantic enrichment for managing and reviewing emergency incident logs

During an emergency incident, several different log files are created by members of the Emergency Response (ER) personnel to document the emergency events that occur throughout the incident. Managing and reviewing these logs is a critical task for understanding and improving the implemented ER actions. A major challenge arising in this task is the merging of log files that are created by the different members of the ER personnel for the incident under study. Extensive manual effort is necessary to identify critical information, such as person names and locations, in order to align and merge the incoming log entries to make them suitable for review. In this paper, we present the WeKnowIt ER Log Manager (WERL), a web-based application that facilitates the task of ER log merging and management by automatically aligning multiple log files and extracting ER-relevant semantic event information from log entry text. WERL makes use of the representation patterns of Event-Model-F in order to facilitate information sharing and reuse. Furthermore, WERL enables interactive exploration of the collected log files by means of temporal, location and semantic filters. Preliminary evaluation of WERL by members of the Sheffield City Council Emergency Planning team confirm that the application provides them with enhanced support during the ER log management and reviewing process.