<![CDATA[The rapid growth of social commerce has intensified competition among online handicraft businesses, making effective store planning increasingly important. While most studies focus on consumer recommendation systems, limited research supports entrepreneurs during the early stage of store configuration. This study proposes a serendipity-aware Decision Support System (DSS) for handicraft store planning using an entropy-weighted hybrid Genetic Algorithm–Particle Swarm Optimization (GA-PSO). A dataset of 105 handicraft stores in West Sumatra was encoded into 19-bit chromosomes representing materials, product types, location, and digital commerce visibility. Entropy-based weighting objectively determined attribute importance without subjective judgment. GA explored store configurations, while PSO optimized evolutionary parameters to balance preference similarity and serendipitous exploration. The proposed framework generated store configurations superior to those in the original dataset. The best solution achieved a preference similarity score of P(x)=0.9108, outperforming the best existing store (P(x)=0.8513) by 6.99%. The hybrid GA-PSO also showed stable performance across multiple runs, indicating robust convergence. This study contributes a data-driven DSS framework integrating entropy weighting, hybrid GA-PSO optimization, and serendipity-aware exploration for entrepreneurial decision support in social commerce.]]>
Copyrights © 2026
