Comparative Research of Swarm Intelligence Clustering Algorithms for Analyzing Medical Data

Abstract

As the Internet of medical Things emerge in the field of medicine, the volume of medical data is expanding rapidly and along with its variety. As such, clustering is an important procedure to mine the vast data. Many swarm intelligence clustering algorithms, such as the particle swarm optimization (PSO), firefly, cuckoo, and bat, have been designed, which can be parallelized to the benefit of mass data computation. However, few studies focus on the systematic analysis of the time complexities, the effect of instances (data size), attributes (dimensionality), number of clusters, and agents of these algorithms. In this paper, we performed a comparative research for the PSO, firefly, cuckoo, and bat algorithms based on both synthetic and real medical data sets. Finally, we conclude which algorithms are effective for the medical data mining. In addition, we recommend the more suitable algorithms that have been developed recently for the different medical data to achieve the optimal clustering. INDEX TERMS Medical data analysis, data mining, swarm intelligence, clustering algorithms.

Swarm Intelligence Algorithms in Text Document Clustering with Various Benchmarks

“…The SI algorithms FFA, CS, BA, and PSO were compared in medical data clustering and results revealed that CS performs slower than others and PSO and BA are relatively faster than others [ 17 ]. Figueiredo et al [ 18 ] studied the comparison of SI algorithms in the clustering domain and concluded that PSO is the best performing algorithm.…”

“…In this study, we focused on comparing the standard SI algorithms. We adopted and used the default parameters of SI algorithms that were introduced in the literature experiments for generic optimization, with different examples [ 17 , 42 ]. Although our experimental problem was text document clustering, which is not an optimization problem, we applied the initial parameter settings of the optimization problem case to ours and tested those standard SI algorithms.…”

“…In the PSO algorithm parameter values were assigned as , , and [ 42 ]. In the BA algorithm parameter values were assigned as , , , and [ 17 ].…”

“…PSO is a simple, fast computing, and parallel processing algorithm [ 43 ]. Comparing other standard SI algorithms, the standard PSO algorithm performs well in clustering and scheduling applications [ 14 , 17 , 18 ].…”

Correction to “Comparative Research of Swarm Intelligence Clustering Algorithms for Analyzing Medical Data”

“…In the above article [1], Zhihua Liu was erroneously omitted as the corresponding author. It should be noted that he should be considered the corresponding author for this article.…”