Recognizing Emotion Cause in Conversations

  title={Recognizing Emotion Cause in Conversations},
  author={Soujanya Poria and Navonil Majumder and Devamanyu Hazarika and Deepanway Ghosal and Rishabh Bhardwaj and Samson Yu and Romila Ghosh and Niyati Chhaya and Alexander F. Gelbukh and Rada Mihalcea},
  journal={Cogn. Comput.},
Recognizing the cause behind emotions in text is a fundamental yet under-explored area of research in NLP. Advances in this area hold the potential to improve interpretability and performance in affect-based models. Identifying emotion causes at the utterance level in conversations is particularly challenging due to the intermingling dynamic among the interlocutors. To this end, we introduce the task of recognizing emotion cause in conversations with an accompanying dataset named RECCON… 

ECPEC: Emotion-Cause Pair Extraction in Conversations

A novel dataset ConvECPE is built and a specifically designed two-step framework for the new ECPEC task, which aims to extract pairs of emotional utterances and corresponding cause utterances in conversations.

Multi-Task Learning and Adapted Knowledge Models for Emotion-Cause Extraction

This work proposes novel methods that combine common-sense knowledge via adapted knowledge models with multi-task learning to perform joint emotion classification and emotion cause tagging and shows performance improvement on both tasks when including common- sense reasoning and a multitask framework.

Multimodal Emotion-Cause Pair Extraction in Conversations

A new task named Multimodal Emotion-Cause Pair Extraction in Conversations, aiming to jointly extract emotions and their associated causes from conversations reflected in multiple modalities, demonstrates the potential of multimodal information fusion for discovering both emotions and causes in conversations.

TSAM: A Two-Stream Attention Model for Causal Emotion Entailment

A Two-Stream Attention Model (TSAM) is proposed to effectively model the speaker’s emotional influences in the conversational history and achieves new State-Of-The-Art (SOTA) performance and outperforms baselines remarkably.

A Consistent Dual-MRC Framework for Emotion-Cause Pair Extraction

A dual-MRC framework to extract emotion-cause pairs in a dual-direction way, which enables a more comprehensive coverage of all pairing cases and outperforms previous methods and achieves state-of-the-art performance.

CARES: CAuse Recognition for Emotion in Suicide Notes

The proposed approaches to emotion-cause identification and extraction are based on pre-trained transformer-based models that attain performance figures of 83.20% ac-curacy and 0.76 Ratcliff-Obershelp similarity, respectively, and suggest that existing computational methods can be adapted to address these challenging tasks.

Position Bias Mitigation: A Knowledge-Aware Graph Model for Emotion Cause Extraction

A novel graph-based method is proposed to explicitly model the emotion triggering paths by leveraging the commonsense knowledge to enhance the semantic dependencies between a candidate clause and an emotion clause and is more robust against adversarial attacks compared to existing models.

Learning a General Clause-to-Clause Relationships for Enhancing Emotion-Cause Pair Extraction

A novel clause-to-clause relationship is described and a general clause-level encoding model named EA-GAT comprising E -GAT and Activation Sort is proposed, which can be broadly integrated with any previous approach.

Perspective-taking and Pragmatics for Generating Empathetic Responses Focused on Emotion Causes

Taking inspiration from social cognition, a generative estimator is used to infer emotion cause words from utterances with no word-level label and a novel method based on pragmatics is introduced to make dialogue models focus on targeted words in the input during generation.

EmoWOZ: A Large-Scale Corpus and Labelling Scheme for Emotion Recognition in Task-Oriented Dialogue Systems

EmoWOZ, a large-scale manually emotion-annotated corpus of task-oriented dialogues based on MultiWOz, is introduced and a novel emotion labelling scheme is proposed, which is tailored to task- oriented dialogues.



Emotion-Cause Pair Extraction: A New Task to Emotion Analysis in Texts

A 2-step approach is proposed to address this new ECPE task, which first performs individual emotion extraction and cause extraction via multi-task learning, and then conduct emotion-cause pairing and filtering.

Detecting Emotion Stimuli in Emotion-Bearing Sentences

A CRF learner is built, a sequential learning model to detect the emotion stimulus spans in emotion-bearing sentences and it is shown that the model significantly outperforms all the baselines.

Emotion Cause Detection with Linguistic Constructions

The multi-label model not only detects multi-clause causes, but also captures the long-distance information to facilitate emotion cause detection and creates two sets of linguistic patterns during feature extraction.

Conditional Causal Relationships between Emotions and Causes in Texts

A new task of determining whether or not an input pair of emotion and cause has a valid causal relationship under different contexts is proposed, and a prediction aggregation module with low computational overhead is proposed to fine-tune the prediction results based on the characteristics of the input clauses.

A Text-driven Rule-based System for Emotion Cause Detection

A text-driven, rule-based approach to emotion cause detection is proposed and a promising system achieves a promising performance for cause occurrence detection as well as cause event detection.

Overview of NTCIR-13 ECA Task

This task is designed to evaluate the emotion cause analysis techniques which automatically determine whether a clause contain the cause of emotion expression, and designed two subtasks: The emotion cause detection subtask for identifying the clause containing the emotioncause corresponding to an emotion expression.

Effective Inter-Clause Modeling for End-to-End Emotion-Cause Pair Extraction

This work tackles emotion-cause pair extraction from a ranking perspective, i.e., ranking clause pair candidates in a document, and proposes a one-step neural approach which emphasizes inter-clause modeling to perform end-to-end extraction.

End-to-End Emotion-Cause Pair Extraction Based on Sliding Window Multi-Label Learning

This work proposes two joint frameworks for ECPE: 1) multi-label learning for the extraction of the cause clauses corresponding to the specified emotion clause (CMLL) and 2)multi-label learn for the extracted emotion clauses corresponding for the specified cause clause (EMLL).

DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset

A high-quality multi-turn dialog dataset, DailyDialog, which is intriguing in several aspects, the language is human-written and less noisy and the dialogues reflect the authors' daily communication way and cover various topics about their daily life.

Joint Learning for Emotion Classification and Emotion Cause Detection

We present a neural network-based joint approach for emotion classification and emotion cause detection, which attempts to capture mutual benefits across the two sub-tasks of emotion analysis.