Humans are commonly seen as the weakest link in corporate information security. This led to a lot of effort being put into security training and awareness campaigns, which resulted in employees being less likely the target of successful attacks. Existing approaches, however, do not tap the full potential that can be gained through these campaigns. On the one hand, human perception offers an additional source of contextual information for detected incidents, on the other hand it serves as information source for incidents that may not be detectable by automated procedures. These approaches only allow a text-based reporting of basic incident information. A structured recording of human delivered information that also provides compatibility with existing SIEM systems is still missing. In this work, we propose an approach, which allows humans to systematically report perceived anomalies or incidents in a structured way. Our approach furthermore supports the integration of such reports into analytics systems. Thereby, we identify connecting points to SIEM systems, develop a taxonomy for structuring elements reportable by humans acting as a security sensor and develop a structured data format to record data delivered by humans. A prototypical human-as-a-security-sensor wizard applied to a real-world use-case shows our proof of concept.
Purpose
In the past, people were usually seen as the weakest link in the IT security chain. However, this view has changed in recent years and people are no longer seen only as a problem, but also as part of the solution. In research, this change is reflected in the fact that people are enabled to report security incidents that they have detected. During this reporting process, however, it is important to ensure that the reports are submitted with the highest possible data quality. This paper aims to provide a process-driven quality improvement approach for human-as-a-security-sensor information.
Design/methodology/approach
This work builds upon existing approaches for structured reporting of security incidents. In the first step, relevant data quality dimensions and influencing factors are defined. Based on this, an approach for quality improvement is proposed. To demonstrate the feasibility of the approach, it is prototypically implemented and evaluated using an exemplary use case.
Findings
In this paper, a process-driven approach is proposed, which allows improving the data quality by analyzing the similarity of incidents. It is shown that this approach is feasible and leads to better data quality with real-world data.
Originality/value
The originality of the approach lies in the fact that data quality is already improved during the reporting of an incident. In addition, approaches from other areas, such as recommender systems, are applied innovatively to the area of the human-as-a-security-sensor.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.