Dialogue Act Recognition via CRF-Attentive Structured Network

Abstract

Dialogue Act Recognition (DAR) is a challenging problem in dialogue interpretation, which aims to attach semantic labels to utterances and characterize the speaker's intention. Currently, many existing approaches formulate the DAR problem ranging from multi-classification to structured prediction, which suffer from handcrafted feature extensions and attentive contextual structural dependencies. In this paper, we consider the problem of DAR from the viewpoint of extending richer Conditional Random Field (CRF) structural dependencies without abandoning end-to-end training. We incorporate hierarchical semantic inference with memory mechanism on the utterance modeling. We then extend structured attention network to the linear-chain conditional random field layer which takes into account both contextual utterances and corresponding dialogue acts. The extensive experiments on two major benchmark datasets Switchboard Dialogue Act (SWDA) and Meeting Recorder Dialogue Act (MRDA) datasets show that our method achieves better performance than other state-of-the-art solutions to the problem. It is a remarkable fact that our method is nearly close to the human annotator's performance on SWDA within 2% gap.

Supervised Clustering of Questions into Intents for Dialog System Applications

“…Finally, our work is related to a large body of research on dialog acts (Stolcke et al, 2000;Kim et al, 2010;Chen et al, 2018): our low-level intent labels (Table 1) can be seen as very finegrained dialog acts (Core and Allen, 1997;Bunt et al, 2010;Oraby et al, 2017).…”

Improved Dynamic Memory Network for Dialogue Act Classification with Adversarial Training

“…Thus, DAs can be thought of as a tag set that classifies utterances according to a combination of pragmatic, semantic, and syntactic criteria. From Table I, we can also find that knowing the past utterances of dialog can help easing the prediction of the current DA state, thus help to narrow the range of utterance generation topics for the current turn [5]. For instance, the "Greeting" and "Farewell" acts are often followed with another same type utterances, the "Answer" act often responds to the former "Question" type utterance.…”

Contextualized Representations for Low-resource Utterance Tagging

“…Moreover, the label assigned to an utterance depends on the current state of the dialogue (Stone, 2005) and prediction of an utterance's label benefits from referring to other utterances in context and their labels (Jaiswal et al, 2019). Deep learning models like RNNs and CNNs have proven effective tools to encode neighbouring utterances (Chen et al, 2018;Liu et al, 2017;Blunsom and Kalchbrenner, 2013;Bothe et al, 2018;Kumar et al, 2017). However such models rely on large annotated corpora that are prohibitively expensive to procure, especially for niche domains.…”