XinChen, Alex Reibman and Sanjay Arora, Ernst & Young LLP, US

Timeliness and contextual accuracy of recommendations are increasingly importantwhen deliveringcontemporary digital marketingexperiences. Conventional recommender systems (RS) suggest relevant but time-invariant items to users by accounting fortheir past purchases. These recommendations only map to customers' general preferences rather than a customer's specific needs immediately preceding a purchase. In contrast, RSs that consider the order of transactions, purchases, or experiencesto measure evolving preferences can offer moresalient and effectiverecommendations to customers:Sequential RSs not only benefit from a better behavioral understanding of a user's current needs but also providebetter predictive power. In this paper, we demonstrate and rank the effectiveness of a sequential recommendation system by utilizing a production dataset of over 2.7 million credit card transactions for 46K cardholders. The methodfirst employs an autoencoder on raw transaction data and submits observed transaction encodings to a GRU-based sequential model. The sequential model produces a MAP@1 metric of 47% on the out-of-sample test set, in line with existing research. We also discuss implications for embedding real-timepredictions using the sequential RS into Nexus, a scalable, low-latency, event-based digital experience architecture.

Sequential Recommendation System, Transaction Data, MLArchitecture, Sequential Neural Network, Auto-encoder, Information Retrieval