Foundational Data Structures: The Frameworks Driving AI Intelligence

Abstract

Data structures like stacks, queues, graphs, and trees have a fundamental foothold in artificial intelligence (AI), providing the necessary scaffolding for efficient computation, organisation of data, and the very invention of algorithms. Those structures empower important processes in AI, such as decision-making, knowledge representation, design of neural networks, and project handling. For knowledge graphs, recommendation systems, and graph neural networks, graphs are employed to model relationships, while trees represent the fundamental paradigms for algorithms pertaining to decision-making, such as decision trees, random forests, and hierarchical clustering. Stacks and queues serve other primitive purposes, including backtracking, parsing, and task scheduling, where these functions find relevant application areas in natural language processing and reinforcement learning. This review scrutinises the effect of these data structures in the field of AI, their development, and how they have adapted to deal with modern challenges, such as scalability, distributed computing, and real-time analytics. Recent progress in hybrid and distributed data structures has also been elaborately explained in this perspective of improving the performance of AI. Refinement of such structures is required more than ever for the applications of emerging AI, such as federated learning, explainable AI, and edge computing. A more profound knowledge of these foundational frameworks is required for elevating AI skills and prompting future innovations.