https://fujeas.fui.edu.pk/index.php/fujeas/issue/feed 2026-03-09T06:33:38+00:00 Dr. Shariq Hussain editor.fujeas@fui.edu.pk Open Journal Systems https://fujeas.fui.edu.pk/index.php/fujeas/article/view/983 2026-03-09T06:33:38+00:00 Farah Haneef farah.haneef@cust.edu.pk Ali Hassan BSE213034@cust.pk Nouman Noor nouman.noor@isb.nu.edu.pk

<p>Automaton learning (Al) is one alternative way to think about artificial intelligence (AI). It provides a structured and comprehensible framework for modeling sequential and structural data. The ability to deduce hidden structures from observed behavior is crucial in a variety of fields, including bioinformatics, software verification, pattern recognition, and natural language processing. Notwithstanding these developments, a large portion of the current research has concentrated on theoretical underpinnings or specific applications, leaving a substantial knowledge vacuum regarding its wider influence across several AI areas. Specifically, the promise of automata learning to solve issues of scalability, generalization, and interpretability is limited by its understudied integration with contemporary machine learning techniques, such as deep learning and reinforcement learning. This paper attempts to close this gap by methodically examining automata learning's function in important AI domains and examining how its use can improve explainability, accuracy, and efficiency. We illustrate that automata learning not only improves on current AI models but also offers a way to create more transparent and dependable intelligent systems by looking at case studies and experimental data. In order to advance next-generation AI, where interpretability and flexibility are just as important as predictive accuracy, the impact assessment indicate that automaton learning's potential must be fully realized.</p>

2026-03-30T00:00:00+00:00 Copyright (c) 2026 Foundation University Journal of Engineering and Applied Sciences