Data mining framework based on rough set theory to improve location selection decisions: A case study of a restaurant chain

Results indicated that latent knowledge can be identified to support location selection decisions, and the proposed data mining framework consists of four stages: problem definition and data collection; RST analysis; rule validation; and knowledge extraction and usage.

Abstract

Location selection plays a crucial role in the retail and service industries. A comprehensive location selection model and appropriate analytical technique can improve the quality of location decisions, attracting more customers and substantially impacting market share and profitability. This study developed a data mining framework based on rough set theory (RST) to support location selection decisions. The proposed framework consists of four stages: (1) problem definition and data collection; (2) RST analysis; (3) rule validation; and (4) knowledge extraction and usage. An empirical study focused on a restaurant chain to demonstrate the validity of the proposed approach. Twenty location variables relevant to five location aspects were examined, and the results indicated that latent knowledge can be identified to support location selection decisions.