[go: up one dir, main page]
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 

Saved Queries

Sign in to use this feature.

Search Filter Reset All

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
Add Subjects Reset
Select Subjects

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
Add Journals Reset
Select Journals

Article Types

remove_circle_outline
Add Article Types Reset
Select Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
Add Countries / Regions Reset
Select Countries / Regions
Update Search
share announcement

Search Results (5)

Search Parameters:
Keywords = emotion–cause pair extraction

Order results
Result details
Results per page
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
16 pages, 728 KiB  
Article
A New Model for Emotion-Driven Behavior Extraction from Text
by Yawei Sun, Saike He, Xu Han and Ruihua Zhang
Appl. Sci. 2023, 13(15), 8700; https://doi.org/10.3390/app13158700 - 27 Jul 2023
Viewed by 1566
Abstract
Emotion analysis is currently a popular research direction in the field of natural language processing. However, existing research focuses primarily on tasks such as emotion classification, emotion extraction, and emotion cause analysis, while there are few investigations into the relationship between emotions and [...] Read more.
Emotion analysis is currently a popular research direction in the field of natural language processing. However, existing research focuses primarily on tasks such as emotion classification, emotion extraction, and emotion cause analysis, while there are few investigations into the relationship between emotions and their impacts. To address these limitations, this paper introduces the emotion-driven behavior extraction (EDBE) task, which addresses these limitations by separately extracting emotions and behaviors to filter emotion-driven behaviors described in text. EDBE comprises three sub-tasks: emotion extraction, behavior extraction, and emotion–behavior pair filtering. To facilitate research in this domain, we have created a new dataset, which is accessible to the research community. To address the EDBE task, we propose a pipeline approach that incorporates the causal relationship between emotions and driven behaviors. Additionally, we adopt the prompt paradigm to improve the model’s representation of cause-and-effect relationships. In comparison to state-of-the-art methods, our approach demonstrates notable improvements, achieving a 1.32% improvement at the clause level and a 1.55% improvement at the span level on our newly curated dataset in terms of the F1 score, which is a commonly used metric to measure the performance of models. These results underscore the effectiveness and superiority of our approach in relation to existing methods. Full article
(This article belongs to the Special Issue Machine Learning and AI in Intelligent Data Mining and Analysis)
►▼ Show Figures

Figure 1

Figure 1
<p>Comparison of ECE, ECPE, and EDBE tasks with an example.</p> Full article ">Figure 2
<p>Overview of the proposed method. It consists of three main parts: an emotion extractor, a behavior extractor, and an emotion–behavior pair filter.</p> Full article ">Figure 3
<p>Structures of the encoder and decoders in the proposed method: (<b>a</b>) BERT encoder; (<b>b</b>) Bi-LSTM decoder for emotion and behavior extractors; and (<b>c</b>) FC decoder for the emotion–behavior filter.</p> Full article ">Figure 4
<p>An example that shows the steps of generating the prompt (<span class="html-italic">P</span>).</p> Full article ">
13 pages, 1367 KiB  
Article
A Token Classification-Based Attention Model for Extracting Multiple Emotion–Cause Pairs in Conversations
by Soyeop Yoo and Okran Jeong
Sensors 2023, 23(6), 2983; https://doi.org/10.3390/s23062983 - 9 Mar 2023
Viewed by 1971
Abstract
People exchange emotions through conversations with others and provide different answers depending on the reasons for their emotions. During a conversation, it is important to find not only such emotions but also their cause. Emotion–cause pair extraction (ECPE) is a task used to [...] Read more.
People exchange emotions through conversations with others and provide different answers depending on the reasons for their emotions. During a conversation, it is important to find not only such emotions but also their cause. Emotion–cause pair extraction (ECPE) is a task used to determine emotions and their causes in a single pair within a text, and various studies have been conducted to accomplish ECPE tasks. However, existing studies have limitations in that some models conduct the task in two or more steps, whereas others extract only one emotion–cause pair for a given text. We propose a novel methodology for extracting multiple emotion–cause pairs simultaneously from a given conversation with a single model. Our proposed model is a token-classification-based emotion–cause pair extraction model, which applies the BIO (beginning–inside–outside) tagging scheme to efficiently extract multiple emotion–cause pairs in conversations. The proposed model showed the best performance on the RECCON benchmark dataset in comparative experiments with existing studies and was experimentally verified to efficiently extract multiple emotion–cause pairs in conversations. Full article
(This article belongs to the Special Issue Emotion Sensing and Robotic Emotional Intelligence)
►▼ Show Figures

Figure 1

Figure 1
<p>Examples of RECCON dataset; (<b>a</b>) example of the utterance having different emotions, and (<b>b</b>) example of two or mor causes representing a single emotion.</p> Full article ">Figure 2
<p>Example of the proposed BIO formatted tags.</p> Full article ">Figure 3
<p>Model architecture.</p> Full article ">
18 pages, 3735 KiB  
Article
A Hierarchical Heterogeneous Graph Attention Network for Emotion-Cause Pair Extraction
by Jiaxin Yu, Wenyuan Liu, Yongjun He and Bineng Zhong
Electronics 2022, 11(18), 2884; https://doi.org/10.3390/electronics11182884 - 12 Sep 2022
Cited by 3 | Viewed by 1746
Abstract
Recently, graph neural networks (GNN), due to their compelling representation learning ability, have been exploited to deal with emotion-cause pair extraction (ECPE). However, current GNN-based ECPE methods mostly concentrate on modeling the local dependency relation between homogeneous nodes at the semantic granularity of [...] Read more.
Recently, graph neural networks (GNN), due to their compelling representation learning ability, have been exploited to deal with emotion-cause pair extraction (ECPE). However, current GNN-based ECPE methods mostly concentrate on modeling the local dependency relation between homogeneous nodes at the semantic granularity of clauses or clause pairs, while they fail to take full advantage of the rich semantic information in the document. To solve this problem, we propose a novel hierarchical heterogeneous graph attention network to model global semantic relations among nodes. Especially, our method introduces all types of semantic elements involved in the ECPE, not just clauses or clause pairs. Specifically, we first model the dependency between clauses and words, in which word nodes are also exploited as an intermediary for the association between clause nodes. Secondly, a pair-level subgraph is constructed to explore the correlation between the pair nodes and their different neighboring nodes. Representation learning of clauses and clause pairs is achieved by two-level heterogeneous graph attention networks. Experiments on the benchmark datasets show that our proposed model achieves a significant improvement over 13 compared methods. Full article
(This article belongs to the Special Issue Advanced Machine Learning Applications in Big Data Analytics)
►▼ Show Figures

Figure 1

Figure 1
<p>A toy example of heterogeneous graph composed of word, clause, and pair nodes.</p> Full article ">Figure 2
<p>(<b>a</b>) An overview of HHGAT; (<b>b</b>) node initialization layer; (<b>c</b>) clause node encoding layer; (<b>d</b>) pair node encoding layer.</p> Full article ">Figure 3
<p>Comparison of experimental results on ECE.</p> Full article ">Figure 4
<p>Visualization of word-clause attention.</p> Full article ">Figure 5
<p>Visualization of meta-path-based attention.</p> Full article ">Figure 6
<p>The inter-graph analysis of meta-path-based attention.</p> Full article ">
14 pages, 1040 KiB  
Article
An End-to-End Mutually Interactive Emotion–Cause Pair Extractor via Soft Sharing
by Beilun Wang, Tianyi Ma, Zhengxuan Lu and Haoqing Xu
Appl. Sci. 2022, 12(18), 8998; https://doi.org/10.3390/app12188998 - 7 Sep 2022
Cited by 2 | Viewed by 1654
Abstract
Emotion–cause pair extraction (ECPE), i.e., extracting pairs of emotions and corresponding causes from text, has recently attracted a lot of research interest. However, current ECPE models face two problems: (1) The common two-stage pipeline causes the error to be accumulated. (2) Ignoring the [...] Read more.
Emotion–cause pair extraction (ECPE), i.e., extracting pairs of emotions and corresponding causes from text, has recently attracted a lot of research interest. However, current ECPE models face two problems: (1) The common two-stage pipeline causes the error to be accumulated. (2) Ignoring the mutual connection between the extraction and pairing of emotion and cause limits the performance. In this paper, we propose a novel end-to-end mutually interactive emotion–cause pair extractor (Emiece) that is able to effectively extract emotion–cause pairs from all potential clause pairs. Specifically, we design two soft-shared clause-level encoders in an end-to-end deep model to measure the weighted probability of being a potential emotion–cause pair. Experiments on standard ECPE datasets show that Emiece achieves drastic improvements over the original two-step ECPE model and other end-to-end models in the extraction of major emotional cause pairs. The effectiveness of soft sharing and the applicability of the Emiece framework are further demonstrated by ablation experiments. Full article
(This article belongs to the Special Issue Natural Language Processing (NLP) and Applications)
►▼ Show Figures

Figure 1

Figure 1
<p>The difference between ECE tasks and ECPE tasks. ECE task aims to extract each cause clause provided emotion annotation, while the ECPE task is targeted at extracting all valid pairs of emotion clauses and the corresponding cause clause in an input document. The orange and green parts are emotion clauses, and the blue is a cause clause.</p> Full article ">Figure 2
<p>An illustration of our proposed end-to-end mutually interactive emotion–cause pair extractor. <math display="inline"><semantics> <mrow> <msub> <mi mathvariant="bold-italic">v</mi> <mn>11</mn> </msub> <mo>,</mo> <msub> <mi mathvariant="bold-italic">v</mi> <mrow> <mn>1</mn> <mi>m</mi> </mrow> </msub> <mo>,</mo> <msub> <mi mathvariant="bold-italic">v</mi> <mrow> <mi>n</mi> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <msub> <mi mathvariant="bold-italic">v</mi> <mrow> <mi>n</mi> <mi>m</mi> </mrow> </msub> </semantics></math> denote the word vector sequence. <math display="inline"><semantics> <msub> <mi mathvariant="bold-italic">s</mi> <mn>1</mn> </msub> </semantics></math> and <math display="inline"><semantics> <msub> <mi mathvariant="bold-italic">s</mi> <mi>n</mi> </msub> </semantics></math> are the clause representation as the output of the word-level encoder. Additionally, <math display="inline"><semantics> <mrow> <msubsup> <mi mathvariant="bold-italic">r</mi> <mn>1</mn> <mi>e</mi> </msubsup> <mo>,</mo> <msubsup> <mi mathvariant="bold-italic">r</mi> <mi>n</mi> <mi>e</mi> </msubsup> <mo>,</mo> <msubsup> <mi mathvariant="bold-italic">r</mi> <mn>1</mn> <mi>c</mi> </msubsup> </mrow> </semantics></math> and <math display="inline"><semantics> <msubsup> <mi mathvariant="bold-italic">r</mi> <mi>n</mi> <mi>c</mi> </msubsup> </semantics></math> are the emotion and cause representation of corresponding clauses. <math display="inline"><semantics> <mrow> <msubsup> <mi mathvariant="bold-italic">a</mi> <mn>1</mn> <mi>e</mi> </msubsup> <mo>,</mo> <msubsup> <mi mathvariant="bold-italic">a</mi> <mi>n</mi> <mi>e</mi> </msubsup> <mo>,</mo> <msubsup> <mi mathvariant="bold-italic">a</mi> <mn>1</mn> <mi>c</mi> </msubsup> </mrow> </semantics></math> and <math display="inline"><semantics> <msubsup> <mi mathvariant="bold-italic">a</mi> <mi>n</mi> <mi>c</mi> </msubsup> </semantics></math> denote the probability distribution of the clause being an emotion clause and a cause clause. <math display="inline"><semantics> <msubsup> <mover accent="true"> <mi mathvariant="bold-italic">r</mi> <mo>˜</mo> </mover> <mi>i</mi> <mi>e</mi> </msubsup> </semantics></math> and <math display="inline"><semantics> <msubsup> <mover accent="true"> <mi mathvariant="bold-italic">r</mi> <mo>˜</mo> </mover> <mi>i</mi> <mi>c</mi> </msubsup> </semantics></math> show the emotion-weighted and cause-weighted clause representations, respectively. <math display="inline"><semantics> <mrow> <mo>(</mo> <msubsup> <mover accent="true"> <mi mathvariant="bold-italic">r</mi> <mo>˜</mo> </mover> <mi>i</mi> <mi>e</mi> </msubsup> <mo>,</mo> <msubsup> <mover accent="true"> <mi mathvariant="bold-italic">r</mi> <mo>˜</mo> </mover> <mi>j</mi> <mi>c</mi> </msubsup> <mo>)</mo> </mrow> </semantics></math> represents potential emotion–cause pairs. <math display="inline"><semantics> <msubsup> <mover accent="true"> <mi mathvariant="bold-italic">y</mi> <mo stretchy="false">^</mo> </mover> <mrow> <mi>i</mi> <mi>j</mi> </mrow> <mi>p</mi> </msubsup> </semantics></math> gives the Bernoulli distribution probabilities of potential emotion–cause pairs to be true.</p> Full article ">Figure 3
<p>Implementation details of the weighted representation. The weighted representation of the cause clause is similar to that of the emotion clause, thus it is omitted here. <math display="inline"><semantics> <mrow> <msub> <mi mathvariant="bold-italic">s</mi> <mrow> <mi>i</mi> <mo>−</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <msub> <mi mathvariant="bold-italic">s</mi> <mi>i</mi> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <msub> <mi mathvariant="bold-italic">s</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> </semantics></math> are the clause representation output by the word-level encoder. <math display="inline"><semantics> <mrow> <msubsup> <mi mathvariant="bold-italic">r</mi> <mrow> <mi>i</mi> <mo>−</mo> <mn>1</mn> </mrow> <mi>e</mi> </msubsup> <mo>,</mo> <msubsup> <mi mathvariant="bold-italic">r</mi> <mrow> <mi>i</mi> </mrow> <mi>e</mi> </msubsup> </mrow> </semantics></math> and <math display="inline"><semantics> <msubsup> <mi mathvariant="bold-italic">r</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>e</mi> </msubsup> </semantics></math> are the emotion and cause representations of corresponding clauses.</p> Full article ">Figure 4
<p>Repeated experiments on the determination of the value of <math display="inline"><semantics> <msub> <mi>λ</mi> <mrow> <mi>s</mi> <mi>f</mi> </mrow> </msub> </semantics></math> in different metric. A higher percentage value and a smaller variance indicate better results. It is not appropriate to use weights that are too high or too low.</p> Full article ">Figure 5
<p>A case study of our method. Our method also works well in extracting emotion–cause pairs when the input text contains multiple emotions and multiple causes that match each other.</p> Full article ">Figure 6
<p>(<b>a</b>) F1 score for different soft-sharing settings in emotion–cause pair extraction tasks. Soft sharing of first layer parameters is far better than not sharing. (<b>b</b>) A detailed comparison of the performance of soft sharing one layer and sharing all layers under different metrics. More layers of soft-sharing parameters do not directly lead to better results.</p> Full article ">
13 pages, 9362 KiB  
Article
Deep Neural Networks Based on Span Association Prediction for Emotion-Cause Pair Extraction
by Weichun Huang, Yixue Yang, Zhiying Peng, Liyan Xiong and Xiaohui Huang
Sensors 2022, 22(10), 3637; https://doi.org/10.3390/s22103637 - 10 May 2022
Cited by 4 | Viewed by 2164
Abstract
The emotion-cause pair extraction task is a fine-grained task in text sentiment analysis, which aims to extract all emotions and their underlying causes in a document. Recent studies have addressed the emotion-cause pair extraction task in a step-by-step manner, i.e., the two subtasks [...] Read more.
The emotion-cause pair extraction task is a fine-grained task in text sentiment analysis, which aims to extract all emotions and their underlying causes in a document. Recent studies have addressed the emotion-cause pair extraction task in a step-by-step manner, i.e., the two subtasks of emotion extraction and cause extraction are completed first, followed by the pairing task of emotion-cause pairs. However, this fail to deal well with the potential relationship between the two subtasks and the extraction task of emotion-cause pairs. At the same time, the grammatical information contained in the document itself is ignored. To address the above issues, we propose a deep neural network based on span association prediction for the task of emotion-cause pair extraction, exploiting general grammatical conventions to span-encode sentences. We use the span association pairing method to obtain candidate emotion-cause pairs, and establish a multi-dimensional information interaction mechanism to screen candidate emotion-cause pairs. Experimental results on a quasi-baseline corpus show that our model can accurately extract potential emotion-cause pairs and outperform existing baselines. Full article
(This article belongs to the Topic Data Analytics and Machine Learning in Artificial Emotional Intelligence)
►▼ Show Figures

Figure 1

Figure 1
<p>Document instances in the dataset.</p> Full article ">Figure 2
<p>The general framework SAP-ECPE for ECPE tasks is introduced. The model consists of three parts, namely span representation, span pairing, and joint prediction, where Emo represents the prediction of the emotion clause and Cau represents the prediction of the cause-clause, span pairing represents the span association pairing module; and joint prediction represents the multidimensional information joint prediction module.</p> Full article ">Figure 3
<p>The processing based on span representation is described in detail, where <span class="html-italic">S</span> represents the sentence representation vector output by Bi-GRU.</p> Full article ">Figure 4
<p>Precision, recall, and F1 value variation of ECPE tasks across different spans.</p> Full article ">Figure 5
<p>F1 changes of emotion clause extraction task and cause clause extraction task in different spans.</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying article 1-50 on page 1 of 1.
Back to TopTop