Dependency networks extractions from textual sources — A case study in criminology

Abstract

The identification and assessment of data is one of the crucial challenges within any scientific discipline. Due to the size, complexity of data, and its internal contradictory information, there are several challenges, which need to be fully addressed to ensure the advance of Data Science and its applications. This work focuses on an automatic approach to extract, identify and discover knowledge focusing on the creation on Dependency Networks (DN). These are powerful modelling tools to navigate throughout data and allow to determine how concepts influence one another. The main motivation of this work is to propose a method to facilitate the decision making and knowledge discovery process. The validation of the proposed approach will demonstrate the potential of this work, specifically focussing on Criminology.