Artificial intelligent environments: Risk management and quality assurance implementation. (2nd January 2020)

• Record Type: Journal Article
• Authors: Malik, Vinita; Singh, Sukhdip
• Is Part Of: Journal of discrete mathematical sciences & cryptography. Volume 23:Number 1(2020)
• Page Start: 187
• Page End: 195
• Publication Date: 2020-01-02
• Subjects: 97R50; (Artificially Intelligent) -- Risk Mitigation -- Fuzzy Logic -- Benchmarking -- Quality Assurance (QA) -- Artificial Neural Networks (ANN) -- Bayesian Regulation Technique; Computer science -- Mathematics -- Periodicals; Cryptography -- Periodicals; Computer science -- Mathematics; Cryptography; Periodicals; 004.0151
• DOI: 10.1080/09720529.2020.1721883
• Languages: English
• ISSNs: 0972-0529

Abstract

Abstract: This research not only elaborates literature on AI techniques for risks management but also proposes a new Neuro-Fuzzy workflow process for mitigating identified software risks. This Neuro-Fuzzy process considers high accuracy levels by training the system with Bayesian regulation approach and Fuzzy logic to construct the fuzzy inference system. The proposed workflow may be implemented by MATLAB or Java Frameworks. The authors have implemented the benchmarking of AI softwares for quantitative performance analysis of the competing products as per technology. The intelligent vendor collects all the metrices of the application by scanning its code and donates them anonymously to its customers across the globe. The organizations use these metrices to benchmark their application against particular technology, application properties and the business drivers. The implementation results of the benchmarking process assists in software robustness and defect probability estimation to assure the software quality. The paper contributes in providing a strong foundation for developing new models, techniques and tools for software quality assurance in AI environments by establishing correlation among risk management and quality assurance in AI environments .