Identifying Potential Instagram Influencers for E-commerce through a Hybrid Approach of Social Network Analysis and Online Sales Behavior

In the contemporary digital landscape, businesses and brands are amplifying online marketing efforts due to the surge in consumerism. Influencer marketing emerges as a potent strategy, leveraging individuals across diverse online channels. As consumers increasingly rely on social media reviews, influencer marketing seamlessly integrates with the decision-making process. This is evident on platforms like Instagram, TikTok, and Facebook, where visual content enhances the consumer experience. Despite the ubiquity of influencer marketing, brands face challenges in identifying suitable influencers for their diverse range of products. Similarly, influencers may lack insights into their followers’ online shopping behaviors, impeding their ability to achieve sales goals. To address these challenges, we propose a novel centrality measurement known as the Socio-Behavioral Network Model. This model incorporates two customizable parameters: Online Sale Behavior Measurement and Network Centrality Measurement. By utilizing data from the Instagram network and E-commerce sources, our model generates scores for micro-influencers, specifically targeting the identification of influencers for E-commerce with the potential to drive product sales. In our experiments, the model achieved an output score of $\sim 53 \%$ using Mean Average Precision at K, demonstrating its capability to successfully identify influencers for E-commerce.