Application of Improved AprioriSome Algorithm in Supermarket O2O Marketing

Abstract

Sequential pattern mining is the key technology for analyzing data. Using Python language and its IDE tool PyCharm can effectively mine the transaction data set generated by supermarket O2O marketing. In this paper, the existing AprioriSome algorithm is improved, and the constraints such as time interval and time window are added, and it is applied to the real transaction data set of a large supermarket chain in Henan. The results show that the running time of the improved AprioriSome algorithm is reduced, and the number of frequent sequences excavated is obviously increased and more practical.