Representation of Data Association Problems using Temporal Relational Probability Models: An Overview

Navdeep Kumar, Meenakshi Jaiswal
International Research Journal of Engineering and Technology   unpublished
Keeping track of dynamic changing complex world is one of the biggest challenges in developing automated systems because of the data association problems. In dynamic world, data association problems arise due to two types of uncertainties i.e. (i) Existence Uncertainty where it is very difficult to predict the actual existence of the objects in the world, (ii) Identity uncertainty in which it becomes very difficult to correlate the identity of an object to its data values. In the present
more » ... o most of the research is being carried out in developing automated systems based on prepositional reasoning (as in case all types of Bayesian networks). These types of prepositional models are not capable enough to represent the complexities of the real world as these fail to establish relations among various objects of real world existence. To represent objects and their relations with other objects there is a need to define models that are capable of defining relations among real world objects and also its associated data values. This paper defines a novel relational model that combines the probability theory with expressive power of first order logic to represent data association problems of complex world which in turn helps to make sound inference in uncertain environment.