Big Data analysis and insight extraction are critical in contemporary large-scale criminal investigations. Analyzing large sets of information related to criminal activities can assist in the identification of correlations between records which can be utilized as strategic intelligence information for the coordination of law enforcement efforts. A Law Enforcement Agency (LEA) officer tasked with this analysis requires tools that can enable Big Data exploration while being intuitive and without requiring the need for prior knowledge. A common practice of data analysis software is the ability to create dashboards that can provide all the necessary visualizations and filters to support such an analysis. However, when it comes to dashboard-wide data exploration through drill-down or roll-up analysis; review and evaluation of correlations between data records; and creation of collections with information important to an investigation, current solutions fall short. In this paper, we present a system that addresses these limitations by facilitating the visualization and exploration of diverse data sources, allowing for individual or concurrent analysis via user-configured dashboards. Individual data records of interest to the analysis can be reviewed and their correlations evaluated by the user, allowing for the creation of collections of records and their relationships based on formalized types of correlations.

Cross-border collaboration among Law Enforcement Agencies is essential for effective and timely suspect identification, especially when the availability of biometric data varies between agencies. This paper presents a scalable and secure approach for multimodal biometric identification across multiple jurisdictions. Our approach allows Law Enforcement Agencies to combine biometric modalities -facial images, fingerprints, and voice samples- and compare them with collaborating agencies, improving the overall accuracy and effectiveness of suspect identification. By leveraging deep learning models for indexing and comparison, efficient data retrieval was achieved without compromising privacy or security. To ensure the protection of sensitive biometric data, our approach incorporates advanced encryption mechanisms, including Homomorphic Encryption for secure computations and Advanced Encryption Standard (AES encryption) for safeguarding biometric information. Its decentralised architecture allows each Law Enforcement Agency to maintain independent instances of the Deep Learning Indexer and Comparator, minimising risks associated with centralising sensitive data and supporting seamless collaboration between agencies. This approach not only improves the accuracy of suspect identification but also enhances operational efficiency by allowing Law Enforcement Agencies to query and share biometric data securely across borders.

This paper addresses the growing issue of terrorists utilizing the Internet, and particularly the Dark Web market places, with the purpose of fundraising for their illegal activities. It proposes the Visual Analytics (VA) system, an advanced AI-powered tool, in an effort to combat cross-border financing associated with terrorism. The tools focus on semantic concept detection and large-scale visual data indexing, and the ultimate goal is to familiarize end-users, practitioners, and law enforcement investigators with these technologies. This paper discusses prior works concerning the presentation and visualization of Deep Learning results to users, outlines the main objectives, and provides fundamental usage instructions for the VA system.

Personal data is any information that relates to an individual. Before disclosing such data to third parties, data controllers must be aware of the de-anonymization risks associated with their datasets and take appropriate anonymization measures. To carry out such actions, data controllers require tools that can analyze the risks in their datasets while also providing the necessary anonymization methods for addressing those risks. Existing tools of this type are insufficient for handling high-dimensional data as well as visualizing their risks. In this paper, we demonstrate DaRAV (De-anonymization Risk Analysis through Visualizations), a tool that addresses these limitations by providing risk analysis methods for five types of complex, high-dimensional data through interactive visualizations, as well as anonymization methods that allow users to create anonymized versions of their data.

In this paper, we are concerned with the run-time communication decisions which need to be made by an intelligent user interface. We model communication decisions as a decision-making process, where a selection among a set of alternative constituents is needed, in order to satisfy specific design goals. Based on techniques from the domains of multiple-criteria decision making and decision making under doubt, we propose the use of a model which takes into account the significance of each constituent toward the satisfaction of each design goal, as well as the consideration of the decision maker (interface designer) regarding the significance of each design goal. Following the proposed approach, a design strategy can be represented by a design vector, and thus, we can define properties of, and relationships between, different design strategies, based on their respective vectors. A specific example deploying the proposed model is presented, based on data from the relevant literature. © 1997 John Wiley & Sons, Inc.