Dialogue Session Segmentation by Embedding-Enhanced TextTiling

TL;DR

This paper introduces an embedding-enhanced TextTiling method for session segmentation in human-computer conversations, leveraging word embeddings to improve robustness against noisy utterances.

Contribution

The paper proposes a novel embedding-enhanced TextTiling approach that outperforms existing methods in session segmentation tasks.

Findings

Improved segmentation accuracy over TextTiling and MMD methods

Embedding enhancement effectively captures semantic context in noisy conversations

Demonstrated robustness in real-world dialogue datasets

Abstract

In human-computer conversation systems, the context of a user-issued utterance is particularly important because it provides useful background information of the conversation. However, it is unwise to track all previous utterances in the current session as not all of them are equally important. In this paper, we address the problem of session segmentation. We propose an embedding-enhanced TextTiling approach, inspired by the observation that conversation utterances are highly noisy, and that word embeddings provide a robust way of capturing semantics. Experimental results show that our approach achieves better performance than the TextTiling, MMD approaches.