ASGARD - Analysis System For Gathered Raw Data
pASGARD has a singular goal, contribute to Law Enforcement Agencies Technological Autonomy and effective use of technology. Technologies will be transferred to end users under an open source scheme focusing on Forensics, Intelligence and Foresight (Intelligence led prevention and anticipation). ASGARD will drive progress in the processing of seized data, availability of massive amounts of data and big data solutions in an ever more connected world. New areas of research will also be addressed. The consortium is configured with LEA end users and practitioners ”pulling” from the Research and Development community who will ”push” transfer of knowledge and innovation. A Community of LEA users is the end point of ASGARD with the technology as a focal point for cooperation (a restricted open source community). In addition to traditional Use Cases and trials, in keeping with open source concepts and continuous integration approaches, ASGARD will use Hackathons to demonstrate its results. Vendor lock-in is addressed whilst also recognising their role and existing investment by LEAs. The project will follow a cyclical approach for early results. Data Set, Data Analytics (multimodal/ multimedia), Data Mining and Visual Analytics are included in the work plan. Technologies will be built under the maxim of ”It works” over ”It’s the best”. Rapid adoption/flexible deployment strategies are included. The project includes a licensing and IPR approach coherent with LEA realities and Ethical needs. ASGARD includes a comprehensive approach to Privacy, Ethics, Societal Impact respecting fundamental rights. ASGARD leverages existing trust relationship between LEAs and the research and development industry, and experiential knowledge in FCT research. ASGARD will allow its community of users leverage the benefits of agile methodologies, technology trends and open source approaches that are currently exploited by the general ICT sector and Organised Crime and Terrorist organisations.
- WG Keim (Data Analysis and Visualization)
|(2021): Visual Analytics for Temporal Hypergraph Model Exploration IEEE Transactions on Visualization and Computer Graphics. IEEE. 2021, 27(2), pp. 550-560. ISSN 1077-2626. eISSN 1941-0506. Available under: doi: 10.1109/TVCG.2020.3030408
Many processes, from gene interaction in biology to computer networks to social media, can be modeled more precisely as temporal hypergraphs than by regular graphs. This is because hypergraphs generalize graphs by extending edges to connect any number of vertices, allowing complex relationships to be described more accurately and predict their behavior over time. However, the interactive exploration and seamless refinement of such hypergraph-based prediction models still pose a major challenge. We contribute Hyper-Matrix, a novel visual analytics technique that addresses this challenge through a tight coupling between machine-learning and interactive visualizations. In particular, the technique incorporates a geometric deep learning model as a blueprint for problem-specific models while integrating visualizations for graph-based and category-based data with a novel combination of interactions for an effective user-driven exploration of hypergraph models. To eliminate demanding context switches and ensure scalability, our matrix-based visualization provides drill-down capabilities across multiple levels of semantic zoom, from an overview of model predictions down to the content. We facilitate a focused analysis of relevant connections and groups based on interactive user-steering for filtering and search tasks, a dynamically modifiable partition hierarchy, various matrix reordering techniques, and interactive model feedback. We evaluate our technique in a case study and through formative evaluation with law enforcement experts using real-world internet forum communication data. The results show that our approach surpasses existing solutions in terms of scalability and applicability, enables the incorporation of domain knowledge, and allows for fast search-space traversal. With the proposed technique, we pave the way for the visual analytics of temporal hypergraphs in a wide variety of domains.
|(2019): Visual Analytics Framework for the Assessment of Temporal Hypergraph Prediction Models Proceeedings of the Set Visual Analytics Workshop at IEEE VIS 2019. 2019
Members of communities often share topics of interest. However, usually not all members are interested in all topics, and participation in topics changes over time. Prediction models based on temporal hypergraphs that—in contrast to state-of-the-art models—exploit group structures in the communication network can be used to anticipate changes of interests. In practice, there is a need to assess these models in detail. While loss functions used in the training process can provide initial cues on the model’s global quality, local quality can be investigated with visual analytics. In this paper, we present a visual analytics framework for the assessment of temporal hypergraph prediction models. We introduce its core components: a sliding window approach to prediction and an interactive visualization for partially fuzzy temporal hypergraphs.
|(2019): Visual Analytics of Conversational Dynamics VON LANDESBERGER, Tatiana, ed., Cagatay TURKAY, ed.. EuroVis Workshop on Visual Analytics (EuroVA). Genf: The Eurographics Association, 2019. ISBN 978-3-03868-087-1. Available under: doi: 10.2312/eurova.20191130
Large-scale interaction networks of human communication are often modeled as complex graph structures, obscuring temporal patterns within individual conversations. To facilitate the understanding of such conversational dynamics, episodes with low or high communication activity as well as breaks in communication need to be detected to enable the identification of temporal interaction patterns. Traditional episode detection approaches are highly dependent on the choice of parameters, such as window-size or binning-resolution. In this paper, we present a novel technique for the identification of relevant episodes in bi-directional interaction sequences from abstract communication networks. We model communication as a continuous density function, allowing for a more robust segmentation into individual episodes and estimation of communication volume. Additionally, we define a tailored feature set to characterize conversational dynamics and enable a user-steered classification of communication behavior. We apply our technique to a real-world corpus of email data from a large European research institution. The results show that our technique allows users to effectively define, identify, and analyze relevant communication episodes.
|research funding program
|01.09.2016 – 29.02.2020