Swarm Intelligence in Data Science: Applications, Opportunities and Challenges

Abstract

The Swarm Intelligence (SI) algorithms have been proved to be a comprehensive method to solve complex optimization problems by simulating the emergence behaviors of biological swarms. Nowadays, data science is getting more and more attention, which needs quick management and analysis of massive data. Most traditional methods can only be applied to continuous and differentiable functions. As a set of population-based approaches, it is proven by some recent research works that the SI algorithms have great potential for relevant tasks in this field. In order to gather better insight into the utilization of these methods in data science and to provide a further reference for future researches, this paper focuses on the relationship between data science and swarm intelligence. After introducing the mainstream swarm intelligence algorithms and their common characteristics, both the theoretical and real-world applications in the literature which utilize the swarm intelligence to the related domains of data analytics are reviewed. Based on the summary of the existing works, this paper also analyzes the opportunities and challenges in this field, which attempts to shed some light on designing more effective algorithms to solve the problems in data science for real-world applications.

A novel firefly algorithm approach for efficient feature selection with COVID-19 dataset

“… [36] suggested the grasshopper algorithm for feature selection, and obtained excellent results on the intrusion detection datasets. It is also worth mentioning that swarm intelligence has also been adopted for other machine learning challenges, specifically hyperparameters’ optimization [22] , [37] , artificial neural network (ANN) training [38] , as well as in for many others [39] .…”

Adaptive Coordinated Motion Control for Swarm Robotics Based on Brain Storm Optimization

“…The statistical results are presented in Figure 10. We test the method 10 times for each population configuration, i.e., 5,7,9,11,13,15,17,19,21,31,41 and 51 respectively. Each bar set in the figure indicates the average errors after convergence, as well as the average convergence time.…”

“…Through a series of iterative operations of convergence and divergence, it can find a relatively optimal solution to a specific problem. The BSO has been widely used in many aspects such as data science [21], electric power systems [22], optimal control [23], computer vision applications [24], wireless sensor networks [25], electromagnetic design problems arXiv:2105.13111v1 [cs.RO] 27 May 2021 [26], as well as multi-robot systems [27]. By applying it to the PID based leader-follower control mechanism, the control parameters will be continuously refined online as the system runs.…”