Improving Residential Marketing Campaigns via Customer Data Clustering

Authors : Mohammed Muqbel, Selçuk Özcan, Çağrı Sel

Pages : 129-144

Doi:10.53433/yyufbed.1463691

View : 77 | Download : 92

Publication Date : 2025-04-29

Article Type : Research Paper

Abstract :As the construction industry struggles to develop effective marketing plans for residential projects, using rich datasets to understand customer demand helps builders of residential complexes with complex use cases. Decision-makers often struggle to understand big data. Solving this problem begins with relevant data being mined and collected. Multi-criteria decision-making (MCDM) models are used to rank the data or alternative options according to their importance to decision-makers. The weights of the criteria, obtained according to their importance, are essential to reveal the relative value of different criteria. To understand and analyze big data, cluster analysis within the data mining discipline is used to segment and score data. This analysis is an effective tool for determining marketing strategies and understanding customer behavior. This study was conducted to determine the marketing strategy appropriate for customer segmentation. For this, the Rank Order Centroid (ROC) criterion weight method gives weights to the criteria according to their relative importance. The K-means cluster analysis algorithm uses the values obtained from the ROC method. By combining ROC and K-means methods, this study will contribute to extracting information from large data sets and simplifying decision-making processes in the residential sector. As a result of the study, customers were divided into groups, and it was concluded that the groups with the highest scores should be prioritized in marketing strategies.
Keywords : Çok kriterli karar verme, Konut, Kümeleme, Pazarlama, Veri madenciliği