"What can I cook with these ingredients?" -- Understanding cooking-related information needs in conversational search

TL;DR

This paper investigates users' cooking-related information needs in conversational search, creating a taxonomy and testing classification models to predict these needs, highlighting the importance of context for system assistance.

Contribution

It introduces a detailed taxonomy of cooking-related information needs and evaluates BERT-based models for predicting these needs during dialogue.

Findings

Achieved 40% F1 score in multi-label classification of needs.

Identified need types that are difficult to predict due to limited context.

Showed that more context improves prediction and system assistance.

Abstract

As conversational search becomes more pervasive, it becomes increasingly important to understand the user's underlying information needs when they converse with such systems in diverse domains. We conduct an in-situ study to understand information needs arising in a home cooking context as well as how they are verbally communicated to an assistant. A human experimenter plays this role in our study. Based on the transcriptions of utterances, we derive a detailed hierarchical taxonomy of diverse information needs occurring in this context, which require different levels of assistance to be solved. The taxonomy shows that needs can be communicated through different linguistic means and require different amounts of context to be understood. In a second contribution we perform classification experiments to determine the feasibility of predicting the type of information need a user has during a dialogue using the turn provided. For this multi-label classification problem, we achieve average F1 measures of 40% using BERT-based models. We demonstrate with examples, which types of need are difficult to predict and show why, concluding that models need to include more context information in order to improve both information need classification and assistance to make such systems usable.