Main Predicate and Their Arguments as Explanation Signals For Intent Classification

TL;DR

This paper introduces a novel method to automatically generate explanation signals for intent classification by marking main predicates and their arguments, creating a new explainability dataset and improving model reasoning.

Contribution

The authors propose a technique to augment intent datasets with explanation signals based on main predicates, and demonstrate its effectiveness in enhancing explainability metrics.

Findings

Models excel at classification but struggle with explainability.

Guiding models with explanation signals improves plausibility scores.

Created a new 21k-instance explainability dataset for intent classification.

Abstract

Intent classification is crucial for conversational agents (chatbots), and deep learning models perform well in this area. However, little research has been done on the explainability of intent classification due to the absence of suitable benchmark data. Human annotation of explanation signals in text samples is time-consuming and costly. However, from inspection of data on intent classification, we see that, more often than not, the main verb denotes the action, and the direct object indicates the domain of conversation, serving as explanation signals for intent. This observation enables us to hypothesize that the main predicate in the text utterances, along with the arguments of the main predicate, can serve as explanation signals. Leveraging this, we introduce a new technique to automatically augment text samples from intent classification datasets with word-level explanations. We mark main predicates (primarily verbs) and their arguments (dependency relations) as explanation signals in benchmark intent classification datasets ATIS and SNIPS, creating a unique 21k-instance dataset for explainability. Further, we experiment with deep learning and language models. We observe that models that work well for classification do not perform well in explainability metrics like plausibility and faithfulness. We also observe that guiding models to focus on explanation signals from our dataset during training improves the plausibility Token F1 score by 3-4%, improving the model's reasoning.