Abstract
This paper explores the integration of artificial neural networks (ANNs) with ontology-based knowledge bases to address complex challenges across various domains. The convergence of machine learning approaches with semantic data representation enables improved monitoring, understanding, and interpretation of intelligent systems that process big data. We examine how integrated approaches facilitate semantic data description processed by ANNs, including the specification of input and output data, and the representation of ANN properties and states. Our research demonstrates the effectiveness of these integrated systems in diverse applications, including electricity demand estimation, solar radiation prediction, location-based services, deep drawing process optimization, and financial market forecasting. Additionally, we analyse the implementation of The VIKOR method is a multi-criteria decision-making approach, for evaluating alternatives when faced with conflicting criteria. The combination of VIKOR
with other techniques such as AHP and fuzzy approaches provides enhanced decision support capabilities. This Combining neural networks with knowledge-based systems provides substantial benefits in terms of performance, flexibility, and interpretability, addressing the limitations of isolated AI development. Our findings contribute to advancing robust AI systems that can effectively handle complex real-world problems
while maintaining interpretability.