Directed Acyclic Graph Network for Conversational Emotion Recognition

Shen, Weizhou; Wu, Siyue; Yang, Yunyi; Quan, Xiaojun

doi:10.18653/v1/2021.acl-long.123

Cited by 144 publications

(85 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ConGCN [13] on the other hand also takes speakers as vertices in addition to individual utterances, which makes it consider the whole ER dataset as a single graph. DAG-ERC [14] uses a Directed Acyclic Graph which combines the benefits of graph and recurrence models with it's structural properties. Moreover, DAG-ERC also makes meaningful and reasonable assumptions while constructing the graph by 1) Removing the link of an utterance in a dialogue to future utterances and 2) By imputing remote information for modeling conversational context by introducing another edge to the speakers previous utterance.…”

Section: Graph-based Modelsmentioning

confidence: 99%

“…We formulate each conversation Uj as a graph and treat each utterance embedding ui ∈ Uj as a node in the graph. In lines with prior-art [8,14], we use the RoBERTa large transformer model, fine-tuned on the task-specific ERC dataset to extract contextualized features ui ∈ R 1024 for for each individual utterance ei in the dialogue. More precisely, similar to [8], we add a [CLS] token at the beginning of each tokenized utterance we…”

Section: Utterance-level Feature Extractionmentioning

confidence: 99%

“…For encoding discourse relations in a conversation, we use a Graph Attention Network (GAT) [30] with the information flow through layers inspired by [31].…”

Section: Temporal Information Flow In Graph Layersmentioning

confidence: 99%

“…where G 1 is the graph-encoded representation of our utterance output by the first layer in the graph. Our graph information propagation step from the hidden layer information of the text to the graphical encoding is inspired by [31]. For each utterance embedding gi, the attention weights between gi and its predecessors are calculated by using gi's hidden state at the (l − 1)-th layer and the nodes j ∈ Ni in the current (l)-th.…”

Section: Temporal Information Flow In Graph Layersmentioning

confidence: 99%

See 3 more Smart Citations

M-MELD: A Multilingual Multi-Party Dataset for Emotion Recognition in Conversations

Ghosh¹,

Srivastava²,

Umesh³

2022

Preprint

View full text Add to dashboard Cite

The expression of emotions is a crucial part of daily human communication. Modeling the conversational and sequential context has seen much success and plays a vital role in Emotion Recognition in Conversations (ERC). However, existing approaches either model only one of the two or employ naive late-fusion methodologies to obtain final utterance representations. This paper proposes a novel idea to incorporate both these contexts and better model the intrinsic structure within a conversation. More precisely, we propose a novel architecture boosted by a modified LSTM cell, which we call DiscLSTM, that better captures the interaction between conversational and sequential context. DiscLSTM brings together the best of both worlds and provides a more intuitive and efficient way to model the information flow between individual utterances by better capturing long-distance conversational background through discourse relations and sequential context through recurrence. We conduct experiments on four benchmark datasets for ERC and show that our model achieves performance competitive to state-of-the-art and at times performs better than other graph-based approaches in literature, with a conversational graph that is both sparse and avoids complicated edge relations like much of previous work. We make all our codes publicly available on GitHub 1 .

show abstract

Section: Graph-based Modelsmentioning

confidence: 99%

Section: Utterance-level Feature Extractionmentioning

confidence: 99%

“…For encoding discourse relations in a conversation, we use a Graph Attention Network (GAT) [30] with the information flow through layers inspired by [31].…”

Section: Temporal Information Flow In Graph Layersmentioning

confidence: 99%

Section: Temporal Information Flow In Graph Layersmentioning

confidence: 99%

See 2 more Smart Citations

M-MELD: A Multilingual Multi-Party Dataset for Emotion Recognition in Conversations

Ghosh¹,

Srivastava²,

Umesh³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Ghosal et al (2020) use common sense knowledge to learn the interaction of interlocutors. Shen et al (2021) design a directed acyclic neural network for encoding the utterances. Hu et al (2021) propose the DialogueCRN to fully understand the conversational context from a cognitive perspective.…”

Section: Introductionmentioning

confidence: 99%

EmoCaps: Emotion Capsule based Model for Conversational Emotion Recognition

Li¹,

Tang²,

Zhao³

et al. 2022

Preprint

View full text Add to dashboard Cite

Emotion recognition in conversation (ERC) aims to analyze the speaker's state and identify their emotion in the conversation. Recent works in ERC focus on context modeling but ignore the representation of contextual emotional tendency. In order to extract multimodal information and the emotional tendency of the utterance effectively, we propose a new structure named Emoformer to extract multimodal emotion vectors from different modalities and fuse them with sentence vector to be an emotion capsule. Furthermore, we design an end-to-end ERC model called Emo-Caps, which extracts emotion vectors through the Emoformer structure and obtain the emotion classification results from a context analysis model. Through the experiments with two benchmark datasets, our model shows better performance than the existing state-of-the-art models.

show abstract

A multiturn complementary generative framework for conversational emotion recognition

Wang

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Conversational emotion recognition (CER) is a significant task due to its application in human–computer interaction. Existing work treats CER as an utterance‐level classification task without considering that empathic response also reflects contextual emotion understanding. Previous work has proven that accurate recognition of emotions in the dialogue history is helpful to generate high‐fit responses. In this paper, we investigate whether this conclusion is a sufficient and necessary condition. Specifically, we define an auxiliary empathic multiturn dialogue generation (MDG) task to enhance emotion understanding. Correspondingly, we present a Sequence‐to‐Sequence oriented framework that combines CER and MDG in a multitask learning manner to verify the complementarity between the two tasks. First, we use alternate recurrent neural networks to encode the content of historical utterances and represent the states of multiparty emotions, which are used for emotion classification. Second, since most MDG methods ignore the emotional coherence of the dialogue context itself, we use affine transformation to fuse hidden states of content and emotions to initialize the decoder. Finally, at each step of generation, an attention mechanism is used to fuse information from the dialogue history to ensure emotional coherence. The CER results of our models outperform the state‐of‐the‐art on three prevalent emotional dialogue data sets. Further analysis demonstrates the mutual promotion and empathy interpretability between MDG and CER. Furthermore, our framework is scalable for different coding strategies and multimodal fusion. To the best of our knowledge, this is the first work to explore CER from the perspective of empathy through multitask learning with dialogue generation.

show abstract

Directed Acyclic Graph Network for Conversational Emotion Recognition

Cited by 144 publications

References 23 publications

M-MELD: A Multilingual Multi-Party Dataset for Emotion Recognition in Conversations

M-MELD: A Multilingual Multi-Party Dataset for Emotion Recognition in Conversations

EmoCaps: Emotion Capsule based Model for Conversational Emotion Recognition

A multiturn complementary generative framework for conversational emotion recognition

Contact Info

Product

Resources

About