The rapid advancement of artificial intelligence (AI), coupled with the global rollout of 4G and 5G networks, has fundamentally transformed the Big Data landscape, redefining data management and analysis methodologies. The ability to manage and analyze such vast and varied datasets has exceeded the capacity of any individual or organization. This study introduces an enhanced framework that expands upon the traditional four Vs of Big Data—volume, velocity, volatility, and veracity—by incorporating six additional dimensions: value, validity, visualization, variability, volatility, and vulnerability. This comprehensive framework offers a novel and straightforward approach to understanding and addressing the complexities of Big Data in the AI era. This article further explores the use of ‘Big D’, an AI-driven, RAG-based Big Data analytical bot powered by the ChatGPT-4o model (ChatGPT version 4.0). This article’s innovation represents a significant advance in the field, accelerating and deepening the extraction and analysis of insights from large-scale datasets. This will enable us to develop a more nuanced and comprehensive understanding of intricate data landscapes. In addition, we proposed a framework and analytical tools that contribute to the evolution of Big Data analytics, particularly in the context of AI-driven processes.

Nature has always been one of the major sources of inspiration and has shown humans many ways to explore; among the best-known examples is the artificial intelligence swarm algorithm, which copies the group behavior of social animals in nature. The Fitness-Dependent Optimizer (FDO) was presented in 2019. It is a swarm intelligence algorithm that takes its inspiration from the reproductive swarming behavior of honeybees. Since its introduction, FDO has become a very popular topic for researchers, and the literature kept growing with different versions of the algorithm, hybrids, and applications specific to the domain. In accordance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) 2020 reporting standard, this paper presents a systematic review of FDO research conducted from 2019 to 2026. Initially, a total of 330 records were retrieved from six academic databases, but after applying the inclusion criteria, only 44 studies were selected and constituted the review material. Firstly, the survey discusses the biological inspiration and offers the mathematical formulation of FDO; secondly, it builds a two axis classification of the variants and modification strategies; Thirdly, it goes through the FDO applications; Fourthly, it offers a quantitative comparison of FDO with three classical metaheuristics Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Artificial Bee Colony (ABC) for convergence behavior, and exploration and exploitation balance; fifthly, it points out priority research gaps. The most decisive finding of this study is that the replacement of the binary weight factor by a continuous or dynamically oscillating value was, by far, the most impactful modification of FDO. Besides that, FDO's best empirical performance is in constrained engineering optimization. Among the open research challenges are scalability to high dimensions, formal convergence analysis, discrete-domain extensions, and reproducibility of benchmark evaluation.