Conditional Dependencies and Position Fixing

Abstract

Experimental crude datasets are usually processed with statistical methods to obtain rough evaluations of nautical measurements. Taking the observations and rectifying the knowledge on them are not correlated. In modern computer applications, raw datasets are usually exploited in the initial learning phase. At this stage, the available data are explored to extract the necessary parameters required within the scheme of computations. The aim of this study was to undertake the crude data processing problem to extract the conditional dependencies that appear as the most important factors when handling the distorted data. First, I upgraded the traditional structures, which are histograms. The stepwise diagrams feature their uncertain evaluation. I upgraded the hierarchy among the evidence within the data pool and defined the given ranking adequate membership functions. The principles of fuzzy systems justified the use of the bin-to-bin additive approach to obtain the locally injective density functions, which can be perceived of as conditional dependency diagrams that enable the construction of simple belief assignments. The structural combination includes the solution to the position fixing problem.