Logo des Repositoriums

Auflistung nach Schlagwort "Forensics"

1 - 4 von 4
  • Zeitschriftenartikel
    Explaining Artificial Intelligence with Care
    (KI - Künstliche Intelligenz: Vol. 36, No. 2, 2022) Szepannek, Gero; Lübke, Karsten
    In the recent past, several popular failures of black box AI systems and regulatory requirements have increased the research interest in explainable and interpretable machine learning. Among the different available approaches of model explanation, partial dependence plots (PDP) represent one of the most famous methods for model-agnostic assessment of a feature’s effect on the model response. Although PDPs are commonly used and easy to apply they only provide a simplified view on the model and thus risk to be misleading. Relying on a model interpretation given by a PDP can be of dramatic consequences in an application area such as forensics where decisions may directly affect people’s life. For this reason in this paper the degree of model explainability is investigated on a popular real-world data set from the field of forensics: the glass identification database. By means of this example the paper aims to illustrate two important aspects of machine learning model development from the practical point of view in the context of forensics: (1) the importance of a proper process for model selection, hyperparameter tuning and validation as well as (2) the careful used of explainable artificial intelligence. For this purpose, the concept of explainability is extended to multiclass classification problems as e.g. given by the glass data.
  • Konferenzbeitrag
    Introducing DINGfest: An architecture for next generation SIEM systems
    (SICHERHEIT 2018, 2018) Menges, Florian; Böhm, Fabian; Vielberth, Manfred; Puchta, Alexander; Taubmann, Benjamin; Rakotondravony, Noëlle; Latzo, Tobias
    Isolated and easily protectable IT systems have developed into fragile and complex structures over the past years. These systems host manifold, flexible and highly connected applications, mainly in virtual environments. To ensure protection of those infrastructures, Security Incident and Event Management (SIEM) systems have been deployed. Such systems, however, suffer from many shortcomings such as lack of mechanisms for forensic readiness. In this extended abstract, we identify these shortcomings and propose an architecture which addresses them. It is developed within the DINGfest project, on which we report and for which we seek initial feedback from the community.
  • Konferenzbeitrag
    Preparing and guiding forensic crime scene inspections in virtual reality
    (Mensch und Computer 2019 - Tagungsband, 2019) Süncksen, Matthias; Hamester, Frederik; Ebert, Lars; Teistler, Michael
    Computer-based scene reconstruction is a method for answering specific forensic questions in the context of accident or crime scenes. For the resulting 3D reconstruction, the use of virtual reality (VR) technology is a novel presentation form. For the presentation to a prosecutor, the need to put visible content into context awards special significance to the moderator, especially as in a VR presentation the head mounted display (HMD) cuts VR users off from their natural environment. We analyze use cases for the parties involved in the courtroom VR presentation and consider the author, moderator and spectator roles and their corresponding session types for creating, directing and watching the presentation. A prototype system has been implemented to allow for suitable VR interactions for the three roles. An evaluation of the system with 12 participants assuming the role of the spectator yielded positive results with regard to the user experience and utility.
  • Konferenzbeitrag
    Shallow CNNs for the Reliable Detection of Facial Marks
    (BIOSIG 2018 - Proceedings of the 17th International Conference of the Biometrics Special Interest Group, 2018) Zeinstra, Chris; Haasnoot, Erwin
    Facial marks are local irregularities of skin texture. Their type and/or spatial pattern can be used as a (soft) biometric modality in several applications. A key requirement for a biometric system that utilises facial marks is their reliable detection. Detection methods typically use a blob detector followed by heuristic post processing steps to reduce the number of false positives. In this paper, we consider shallow Convolutional Neural Networks (CNNs) for facial mark detection. The choice of this network type seems natural as it learns multiple (non) blob detectors; shallow refers to the fact that we only consider CNNs up to three layers.We show that (a) these CNNs successfully address the false positive problem, (b) remove the need for post processing steps, and (c) outperform a classic blob detector, approaches taken in previous studies and some other non CNN type classifiers in terms of EER and FMR at TMR=0.95.