A hybrid big data analytical approach for analyzing customer patterns through an integrated supply chain network

Abstract

The recent technology innovation such as big data and its applications has been adopted widely in industries in order to deal with massive datasets. Through data integration, data analysis, and data interpretation, big data technologies can assist business stakeholders in gaining the benefits in their decision-making process. In this research, we hypothesize that combining several big data analytical methods for analyzing integrated customer data can provide more effective and intelligent strategies. A hybrid model combining recency, frequency, and monetary value (RFM) model, K-means clustering, Naïve Baye's algorithm, and linked Bloom filters is proposed to target different customer segments. Our results suggest that (1) the use of big data analytics can provide marketers a direction to make marketing strategies; (2) the use of big data analytics can predict potential customer demands; and (3) the proposed linked Bloom filters can store inactive data in a more efficient way for future use.