The problem of missing data in LSP aggregation

Abstract

Aggregation of continuous logic variables or degrees of fuzzy membership using soft computing aggregation models assumes the availability of all input data. Unfortunately, in many applications some inputs are missing. In this paper we propose an aggregation process that tolerates missing data. The aggregation process is implemented in the context of LSP evaluation criteria. Using the proposed method the aggregators automatically reconfigure themselves so that only the available data are aggregated. The corresponding evaluation decisions can be based on incomplete data. © 2012 Springer-Verlag Berlin Heidelberg.