Methodology for Data and Information Quality Assessment in the Context of Emergency Situational Awareness

摘要

AbstractSituation Assessment (SA) approaches aim to provide powerful resources to support decision makers in enhancing their Situational Awareness (SAW). The process of SA in emergency response systems is of utmost importance once the information acquired and inferred from human reports is used to support the deployment of tactics and resources to attend incidents. However, operators of such systems may face informational barriers leading to an erroneous SAW and consequently jeopardize the assessment process if they are not handled. One of such barriers in this context is the presence of low-quality data or information. Hence, a challenging issue in this field is to determine how to generate, score, update and represent data and information quality cues to support operators to reason under uncertainties and improve their understanding about an ongoing situation. The state of the art in this area presents a research gap regarding methodologies for the information quality assessment which can be used in the emergency management domain. Also, there is a lack of approaches that interface with different levels of situational information during an assessment routine. Hence, in order to enhance operators situational awareness, a new methodology is presented to improve the capabilities of SA systems by enriching knowledge about situations with reliable metadata. Such methodology, named Information Quality Assessment Methodology in the Context of Emergency situational awareness, is composed by: elicitation of data and information quality requirements; definition of functions and metrics to quantify quality dimensions, such as completeness, timeliness, consistency, relevance and uncertainty; and the representation of situational information by the instantiation of a situation model, which can be consumed by an ontology. Finally, a case study is addressed to verify the applicability of the methodology using data and information from a robbery event. The results obtained show situational models with qualified information that feed SA systems, enabling them to be aware of information quality.