Lisna lisna Zahrotun
Ahmad Dahlan University

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Random perturbation bee colony optimized k-means approach for optimized MSME data clustering Lisna lisna Zahrotun; Ika Arfiani; Dwi Normawati
International Journal of Advances in Intelligent Informatics Vol 12, No 3 (2026): August 2026
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v12i3.1781

Abstract

Micro, Small, and Medium Enterprises (MSMEs) play an important role in the economy and need strategic support to grow. The K-Means method is often used for cluster analysis, but has a weakness in determining the initial cluster centre. This research proposes the K-Means Random Perturbation Bee Colony Optimisation (RPBCO) method to overcome the problem. The test results show an increase in cluster accuracy, with the Silhouette Coefficient score increasing by 40.9% (from 0.171 to 0.245). Wilcoxon testing also showed a z-value of -1.342, confirming that RPBCO is superior to standard K-Means. This method proved effective in optimising clusters on a heterogeneous MSMEs dataset. The analysis revealed that creative MSMEs thrive on Instagram, while retail MSMEs perform best on Shopee. To expand the market reach of MSMEs beyond Java and internationally, marketing strategies tailored to the characteristics of the region and target market are required. Creative MSMEs can utilise Instagram Ads to target ASEAN countries, while retail MSMEs can focus on global platforms such as Amazon or Alibaba. Digital literacy training, platform algorithm workshops, and collaborative marketing campaigns can strengthen this strategy. These measures are expected to increase turnover and support the sustainable growth of MSMEs, positively impacting more than one million businesses.