With more and more people shopping online, companies deal with customer data input not onlyin high volume but also dynamic. In order to attract target customers more effectively and toprovide customers with more personalized services, how to automatically extract personalpreference from the real-time data and make real-time recommendation has been growing inimportance for businesses in the competitive modern society. Current data analysis methods foronline shopping recommendation largely rely on historical transaction record. Analyses haveindicating that next items a customer would like to buy not only depend on one’s past historicalrecords but on the item currently being put into the shopping cart. This paper designs an engineto combine each customer’s past transaction and current shopping cart data to dynamicallyinfer one’s preference for the next items. The design, Transaction-Data Based Real-timePreference Inference Engine (TRPIE), consists of two innovative ideas. The first exploits thepurchasing sequence information and turns one’s purchase history into a temporal series ofdata, where a customer’s dynamic purchasing behaviour information lies. The second is adesign of a two-layer Recurrent Neural Network (RNN) for extracting personal purchasingpreference pattern from the temporal series of data to infer preference of next items. Areference implementation of TRPIE design integrates existing tools such as Keras, tensorflowTM,sklearnTM, and MlxtendTM. Test results over real data from 1,374 people show that predictionaccuracy has doubled that obtained by a basket analysis method, which ignores sequentiality ofpurchasing items.
展开▼
