Home // IMMM 2014, The Fourth International Conference on Advances in Information Mining and Management // View article

Semantic Mining and Analysis of Heterogeneous Data for Novel Intelligence Insights

Authors:
Rick Adderley
Patrick Seidler
Atta Badii
Marco Tiemann
Federico Neri
Matteo Raffaelli

Keywords: Data Mining; Text Mining; Entity Recognition and Resolution; Social and Criminal Network Analysis; Semantic Interoperability

Abstract:
This paper describes the implementation of a Data Mining (DM) system under the EU FP7 Security Research Project Multi-Modal Situation Assessment & Analytics Platform (MOSAIC). The system aims to enable the part-automatic detection and recognition of crime threats in uncertain environments. It facilitates the automatic retrieval of intelligence data providing deep semantic information access and dynamic classification features for distributed data sources, such as Policing legacy databases, Police text documents and free text database fields. A specific pipeline of linguistic processors that share a common knowledge base on crime patterns has been created to retrieve entities and events from text documents and websites. Structured and unstructured data retrieved from the individual data sources are integrated in a semantically query-able unified data representation using specific ontological models. A domain specific entity resolution module ensures the resolution of conflicting and misleading identities to enable data retrieval and fusion from disparate data sets. As criminal network analysis depicts a major part of the intelligence process, specific measures and algorithms have been developed to support analysts in retrieving, analysing, and disrupting criminal networks.

Pages: 36 to 40

Copyright: Copyright (c) IARIA, 2014

Publication date: July 20, 2014

Published in: conference

ISSN: 2326-9332

ISBN: 978-1-61208-364-3

Location: Paris, France

Dates: from July 20, 2014 to July 24, 2014