Data-Driven Analysis of Social Commerce Features Impact on Purchase Intention Using SEM-PLS

Social commerce platforms integrate interactive social media features with e-commerce functionalities, creating complex data-driven environments that significantly influence consumer purchasing behavior. This study aims to analyze the impact of key social commerce features such as social presence, information quality, and emotional support on consumer purchase intentions, with a particular focus on the mediating roles of perceived risk and trust. Employing the Stimulus-Organism-Response (SOR) theoretical framework, the research collects data from 110 active social commerce users and utilizes Structural Equation Modeling with Partial Least Squares (SEM-PLS) to quantitatively examine the relationships among the constructs. The results demonstrate that social commerce constructs exert both direct and indirect effects on purchase intentions, with perceived risk mitigation emerging as a critical factor driving consumer decisions. Although trust is positively influenced by social commerce features, it does not directly determine purchase intention within the studied context. The findings emphasize the necessity for secure transaction mechanisms, transparent product information, and enhanced user engagement strategies to effectively reduce perceived risk and increase conversion rates. This study contributes a robust analytical framework that supports the development of intelligent social commerce platforms and offers practical implications for system designers and digital marketers aiming to optimize consumer interaction and transaction security. The integration of advanced data-driven risk and trust management techniques underscores the evolving complexity of social commerce ecosystems and highlights opportunities for leveraging technology to enhance user experience and commercial success.