Novel Approach and Innovative Strategy for Mall Customer Segmentation Using Machine Learning Techniques

Dividing up a company’s clientele into distinct groups is one of the most critical components of its decision-making support system. It is an essential piece of marketing tool that enables the targeting of particular client groups through the application of specialized marketing strategies. In most instances, clustering approaches serve as the foundation for client segmentation. This study’s findings have the potential to improve marketing techniques as well as product design. It can be used to improve customer satisfaction, encourage customer participation, enhance the effectiveness of marketing campaigns by participating in targeted marketing activities, and do away with transaction fees associated with merchant agreements that do not significantly influence consumer preferences. A company must identify potential customers in a highly competitive market on time. This will make it possible for the organization to increase its customer base gradually. This research presents an innovative approach to customer segmentation using spectral clustering and clustering ensemble. The possibility exists for our model to evolve into required items for boosting the general clustering solutions quality. It combines the results from several different clustering methods based on spectral clustering. These methods include Mini Batch K-means, K-means, and MeanShif.