A Rule-Based Knowledge Discovery Engine Embedded Semantic Graph Knowledge Repository for Retail Business

Jiandong Li, Atsushi Sato, Runhe Huang, Dongming Chen

Abstract: Retail data available for consumer-oriented company is a precious asset which can deliver useful insights in decision making and marketing strategy. KID (Data-Information-Knowledge) model which is a generic from data to knowledge cognitive model can be applied to retail business for supporting retail data analytics. The core part of KID model is the knowledge repository which should be designed to support interpreting streaming data into meaningful information, assimilating meaningful information and updating knowledge. Based on Grüninger and Fox method which is an approach of rebuilding ontology, Neo4j graph database and retail ontology designed by Maryam Fazel Zarandi, a retail semantic graph knowledge repository embedding a rule-based knowledge discovery engine is proposed and developed for KID model. It can translate data into information and absorb information into graph knowledge repository by pre-embedded prior objective-oriented algorithms knowledge in algorithm pool, deduce answers to retail queries from graph knowledge repository by the deductive reasoning ability provided by Jess rule engine. This paper also shows a customer value assessment case study by using RFM and K-means algorithms to demonstrate the proposed graph knowledge repository.