Using a Language Model in a Kiosk Recommender System at Fast-Food Restaurants

TL;DR

This paper presents a kiosk recommender system for fast-food restaurants that uses a language model and neural network to improve recommendation accuracy, demonstrating superior offline performance and competitive online results.

Contribution

It introduces a novel combination of a language model and neural network classifier for kiosk recommendations in fast-food settings.

Findings

Outperforms other models in offline tests

Achieves comparable performance to best models in online A/B/C tests

Demonstrates effectiveness of language model-based recommendations

Abstract

Kiosks are a popular self-service option in many fast-food restaurants, they save time for the visitors and save labor for the fast-food chains. In this paper, we propose an effective design of a kiosk shopping cart recommender system that combines a language model as a vectorizer and a neural network-based classifier. The model performs better than other models in offline tests and exhibits performance comparable to the best models in A/B/C tests.