Using ML for personalizing food search at Gojek

Description

GoFood, Gojek’s food delivery product, is one of the largest of its kind in the world. Jewel James and Mudit Maheshwari discuss the approaches considered and lessons learned during the design and successful experimentation of a search system that uses ML to personalize the restaurant results based on the user’s food and taste preferences.

The estimation of the relevance was formulated as a learning to rank ML problem, which makes the task of performing the ML inference for a very large number of customer-merchant pairs the next hurdle. Jewel and Mudit outline their findings for creating a learning model for food recommendations, targeting experiments to a certain percentage of users, training the model from real time data, and enriching restaurant data with custom tags.

Join in for tips on making design decisions on the data pipelines and software architecture needed when using ML for relevance ranking in high-throughput search systems.