The semantic interpretation of cognitive models helps to improve the quality of networked group decision-making processes. The consideration of illogical, uncaused factors’ relationships of cognitive models is particular complexity. The present work is aimed constructing an approach of verifying factors and their mutual relationships in the cognitive models on the base of Big Data analyzing under both logical and uncaused factors’ relationships. The paper also shows the possibility of distorting the semantic interpretation of cognitive models because of discrete data processing. The application of resonance-analog mechanisms for constructing semantic interpretations for modeling is substantiated