Tree Based Reduction of Concept Lattices Based on Conceptual Indexes

There are many approaches and tools which deal with conceptual structures in datasets and their main goal is to support user in understanding of data and structure. One of methods is formal concept analysis (FCA) which is suitable for processing and analyzing input data of object-attributes models based on their relationship. One from FCA family is model of generalized one-sided concept lattice (GOSCL). It is suitable to work with different type of attributes. While generating one-sided concept lattices in FCA improved understanding and interpretation of analysis, one of the lasting problem is to provide the users a result of FCA in appropriate form, if there is large number of concept lattices and generated structure is complex. This is one of the main topics in the FCA and solution can be reached with the reduction methods. In this paper we propose some of the reduction techniques and their combinations.