EXPLORING FUZZY GEOGRAPHIC PROFILING THROUGH MVPP AND BMVP APPROACHES

Abstract

A fuzzy extension of Minimal Variance Projection Profiling (MVPP), a geospatial analysis technique for locating possible areas of interest in geographical profiling, is presented in this study. In traditional MVPP, the spatial distribution of criminal events is analyzed using statistical measures, linear algebra, and Euclidean geometry. A minimal variance line and a bounding polygon that is likely to contain an offender's hideout are constructed. By adding fuzzy matrices, fuzzy covariance, and fuzzy distances, we expand MVPP in this fuzzy form to address spatial uncertainty. This method accounts for imprecision in spatial data by treating crime event locations and projections as fuzzy data points with different levels of membership. In situations where there is insufficient or unclear evidence, the fuzzy MVPP framework efficiently captures regions of interest, providing a more adaptable and realistic option for comprehending illegal spatial behavior. Through the provision of a rigorous, fuzzy-based approach for evaluating ambiguous spatial data in criminal investigations, this contribution enhances the field of geographical profiling.