Comparative Analysis for Customer Profiling and Segmentation in Food and Beverages Using Data Mining Techniques

The exponential growth of social media data has been collected in today's F&B business. Simple and cost-effective to run high business performance result is critical and engage in successful marketing campaigns. It is necessary to understand different customer behaviour and make more strategically decisions to explore and analyze these data to target the appropriate customers in their business. Most of the profiling and segmentation consider as popular marketing approach to investigate customers based on their characteristics for targeted campaigns. In this paper, we use different clustering methods to get insight and reveal the trends in the mass of real-time collected datasets. K-Means, agglomerative clustering and farthest clustering algorithm is applied to segment the customers into the similar group and further dive into relating specific subset of given datasets. The result of the comparison study shows K-means clustering achieved the accuracy of seventy-five percent (75%) outperform to others. This study addresses the effective data mining approach has helped the F&B companies to make more branding in their promotion campaign and serve the customer better.