Towards a Unified Approach to Information Integration - A review paper on data/information fusion

Abstract

This document was printed on recycled paper. AbstractData (or information) fusion (or integration) from different sources is found in many applications, from epidemiology, medicine, biology, and business to military applications such as intelligence. Data fusion may involve integration of spectral content with imaging, text, and many other observations or sensor data. In this paper, we review the methodologies and implementations of the data-fusion process used in the literature and illustrate in more detail the methodologies involved by presenting three examples. We propose that the data-fusion process can be viewed as multi-path and multi-stage mappings and that the

Mathematical Fundamentals of Probabilistic Semantics for High-Level Fusion

“…This structure allows for the mapping of specific fusion capabilities to their function and performers. The final area they reviewed was Salerno's conceptual functional flow model [26].…”

“…According to Salerno, over thirty fusion process models had been proposed by 2002[23]. Several teams have reviewed selected subsets, including Esteban et al[24], Bedworth and O'Brien[25], Whitney, Posse and Lei[26] and Roy et al[27]. Kokar, Bedworth and Frankel categorized many of the models by whether they were information centered or process centered.…”