A Unified Multi-perspective Artificial Neural Networks (ANN) Based Performance Evaluation and Its Quantitative Validation Framework

The numerous merits and prospects of Artificial Neural Networks (ANN) in various interdisciplinary disciplines have become part and parcel of the present era of industrial globalization. They comprise a variety of integral components that have a high degree of performance merits to meet high market demands. A new approach to performance evaluation is needed to deal with the cutting-edge technology present in the era of big data. In the book chapter, a variety of concepts, models, and approaches are discussed for assessing performance models at the prototype level. Here, the aim was to explore Artificial Intelligence (AI) techniques to handle big data in conjunction with the appropriate approaches. The model is validated using metrics established on the Benefits-Opportunities-Costs-Risks (BOCR) framework, balancing benefits, and opportunities at the front end and matching them with risk and cost merits at the other end. Performance validation model formalism has been introduced and put forward from a quantitative economic perspective. This model framework is designed to evaluate the performance of complex, distributed, and transactional application problems by addressing non-functional requirements.