Ingredient Extraction from Text in the Recipe Domain

TL;DR

This paper presents a BERT-based model for extracting ingredients from plain-text recipe queries, achieving high accuracy and aiding virtual assistants in understanding user requests.

Contribution

It introduces a fine-tuned BERT model specifically designed for ingredient extraction in the recipe domain, with publicly available code.

Findings

Achieved an F1-score of 95.01% on ingredient extraction.

Demonstrated the effectiveness of BERT in domain-specific information extraction.

Provided open-source code for reproducibility.

Abstract

In recent years, there has been an increase in the number of devices with virtual assistants (e.g: Siri, Google Home, Alexa) in our living rooms and kitchens. As a result of this, these devices receive several queries about recipes. All these queries will contain terms relating to a "recipe-domain" i.e: they will contain dish-names, ingredients, cooking times, dietary preferences etc. Extracting these recipe-relevant aspects from the query thus becomes important when it comes to addressing the user's information need. Our project focuses on extracting ingredients from such plain-text user utterances. Our best performing model was a fine-tuned BERT which achieved an F1-score of $95.01$. We have released all our code in a GitHub repository.