This paper proposes a framework for mining market basket data to generate Nutritional Patterns (NPs) and a method for analysing generated nutritional patterns using Fuzzy Association Rule Mining. Edible attributes are filtered from transactional input data by projections and are then converted to Required Daily Allowance (RDA) numeric values. The RDA database is then converted to a fuzzy database that contains expended normalized fuzzy attributes comprising of different fuzzy sets. Analysis of nutritional information is performed either from normal generated association rules or from a converted fuzzy transactional database. Our approach uses prototype support tool that extract Nutritional Patterns (NPs) and signifies the level of nutritional content in an association rule per item. The paper presents various performance tests and interestingness measures to demonstrate the effectiveness of the approach and concludes with exper...
M. Sulaiman Khan, Maybin K. Muyeba, Frans Coenen