Multi-Granularity Open Intent Classification via Adaptive Granular-Ball Decision Boundary

TL;DR

This paper introduces MOGB, a novel multi-granularity approach for open intent classification that uses adaptive granular-ball decision boundaries to better distinguish known and unknown intents, especially in complex semantic spaces.

Contribution

The paper proposes a hierarchical representation learning method with adaptive granular-ball clustering and multi-granularity decision boundaries for improved open intent classification.

Findings

Effective in capturing fine-grained intent structures

Outperforms existing methods on public datasets

Demonstrates robustness in distinguishing known and unknown intents

Abstract

Open intent classification is critical for the development of dialogue systems, aiming to accurately classify known intents into their corresponding classes while identifying unknown intents. Prior boundary-based methods assumed known intents fit within compact spherical regions, focusing on coarse-grained representation and precise spherical decision boundaries. However, these assumptions are often violated in practical scenarios, making it difficult to distinguish known intent classes from unknowns using a single spherical boundary. To tackle these issues, we propose a Multi-granularity Open intent classification method via adaptive Granular-Ball decision boundary (MOGB). Our MOGB method consists of two modules: representation learning and decision boundary acquiring. To effectively represent the intent distribution, we design a hierarchical representation learning method. This involves iteratively alternating between adaptive granular-ball clustering and nearest sub-centroid classification to capture fine-grained semantic structures within known intent classes. Furthermore, multi-granularity decision boundaries are constructed for open intent classification by employing granular-balls with varying centroids and radii. Extensive experiments conducted on three public datasets demonstrate the effectiveness of our proposed method.