REGRESSION ANALYSIS APPLICATION USING ML.NET

This research has presented the design, implementation, and evaluation of an innovative Dual-Approach Analytics Platform that uniquely bridges the gap between AI-automated and manual ML workflows within a unified system. Built on ML.NET, ASP.NET Core, React, and Claude AI, the platform successfully demonstrates that a single application can serve both novice users and expert data scientists through architectural innovation and professional-grade controls. Key contributions include an AI analysis engine that achieves 93-95% accuracy with zero user configuration, alongside a manual pathway providing granular access to over 15 encoding strategies and 20 algorithms for specialized optimization. User studies validated this design, showing that the dual-user interface serves both audiences without compromise, democratizing machine learning for non-technical users while providing experts with tools that rival code-based frameworks. Ultimately, this platform establishes a new standard in applied ML tooling, proving that automation and professional control can coexist harmoniously to meet diverse organizational analytics needs within an enterprise-grade, secure, and scalable environment.

[1]              M. Feurer et al., “Efficient and Robust Automated Machine Learning,” in Advances in Neural Information Processing Systems (NIPS), 2015, pp. 2962-2970 [2]              Microsoft, "ML.NET: Machine Learning for .NET Developers," Microsoft Documentation, 2023. [Online]. Available:  https://learn.microsoft.com/en-us/dotnet/machine-learning/ [3]              M. B. Jones, J. Bradley, and N. Sakimura, "JSON Web Token (JWT)," RFC 7519, Internet Engineering Task Force (IETF), May 2015. [4]              R. Wieringa et al., "Refresh Token Rotation and Security in OAuth 2.0 Systems," Journal of Web Security, vol. 11, no. 2, pp. 34–51, 2023. [5]              A. Narayan et al., "Can Foundation Models Wrangle Your Data?" Proceedings of VLDB Endowment, vol. 16, no. 4, pp. 738–746, 2022. [6]              A. Satyanarayan et al., "Vega-Lite: A Grammar of Interactive Graphics," IEEE Transactions on Visualization and Computer Graphics, vol. 23, no. 1, pp. 341–350, 2017. [7]              A. Vaswani et al., "Attention Is All You Need," in Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 5998–6008. [8]              T. Chen and C. Guestrin, "XGBoost: A Scalable Tree Boosting System," in Proceedings of ACM SIGKDD, 2016, pp. 78