On inference rules in decision formal contexts
Jinhai Li, Changlin Mei, Lidong Wang, Junhong Wang · International Journal of Computational Intelligence Systems · 2014
Rule acquisition is one of the main purposes in the analysis of decision formal contexts.In general, the number of implications in a decision formal context is an exponential increase to the scale of the database.So, it is important to introduce effective inference rules between implications for eliminating as many superfluous implications as possible.This study puts forward a criterion called 'strongness' to assess the effectiveness of inference rules in terms of eliminating superfluous implications.We define a new inference rule in decision formal contexts and prove that the proposed inference rule is stronger than the existing one.Furthermore, we figure out the exact number of the superfluous implications that we can additionally remove by using the proposed inference rule compared with the existing one.