VALCRI provides a challenging and overwhelming high-dimensional dataset that comprises of hundreds of extracted semantic features in addition to the usual spatiotemporal information or metadata. To overcome the curse of dimensionality and to generate low-dimensional representations of these semantic features we apply interactive high-dimensional data analysis techniques with the goal of obtaining clusters of similar crime reports. However, it is still a challenge for crime analysts to make sense of the results and to provide useful interactive feedback to the system. Therefore, we provide several tightly integrated interactive visualizations that allow the analysts to identify clusters of similar crimes from different perspectives and interactively focus their analysis on features or crime records of particular interest.
Criminal Intelligence, High-Dimensional Data Analysis, Feature Extraction, Dimensionality Reduction, Visual Analytics
VALCRI WHITE PAPER SERIES