It Takes Two To Tango: Modification of Siamese Long Short Term Memory Network with Attention Mechanism in Recognizing Argumentative Relations in Persuasive Essay

Gema, Aryo Pradipta; Winton, Suhendro; David, Theodorus; Suhartono, Derwin; Shodiq, Muhsin; Gazali, Wikaria

doi:10.1016/j.procs.2017.10.036

Cited by 10 publications

(7 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the continuous attempt to limit the number of handcrafted features and corpus-specific knowledge, research has mainly shifted towards Deep Learning. The most common architecture is bidirectional Long-Short Term Memory (LSTM) Neural Networks fed with word embeddings [23,24,7]. In general, Deep Learning frameworks tend to give state-of-the-art results, which approach human performance.…”

Section: Related Work On Argument Identificationmentioning

confidence: 99%

“…Therefore, we aim to address those challenges of AM by using a novel Transfer Learning solution that is inter-domain applicable and does not require any labor-intensive NLP features. Current solutions in AM with unsupervised learning (e.g., [6]) or classification approaches using embedding structures and neural networks ( [7], [8]) fall short of solving those issues, since they are either not generalizable or very domain-specific. Hence, we aim to contribute to literature and practice by presenting a novel solution that works on a Deep Learning model architecture and enables future scientists and researchers to build AM pipelines without intensive effort.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unlocking Transfer Learning in Argumentation Mining: A Domain-Independent Modelling Approach

Wambsganss¹,

Molyndris²,

Söllner³

2020

WI2020 Zentrale Tracks

View full text Add to dashboard Cite

Argument identification is the fundamental block of every Argumentation Mining pipeline, which in turn is a young upcoming field with multiple applications ranging from strategy support to opinion mining and news fact-checking. We developed a model, which is tackling the two biggest practical and academic challenges of the research field today. First, it addresses the lack of corpus-agnostic models and, second, it tackles the problem of human-laborintensive NLP models being costly to develop. We do that by suggesting and implementing an easy-to-use solution that utilizes the latest advancements in natural language Transfer Learning. The result is a twofold contribution: A system that delivers state-of-the-art results in multiple corpora and opens up a new way of academic advancement of the field through Transfer Learning. Additionally, it provides the architecture for an easy-to-use tool that can be used for practical applications without the need for domain-specific knowledge.

show abstract

Section: Related Work On Argument Identificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unlocking Transfer Learning in Argumentation Mining: A Domain-Independent Modelling Approach

Wambsganss¹,

Molyndris²,

Söllner³

2020

WI2020 Zentrale Tracks

View full text Add to dashboard Cite

show abstract

“…It was done by observing either the relation was sufficient or not [11]. Long Short Term Memory (LSTM) as one of promising deep learning method for text was modified involving Siamese network to recognize argumentation relation in persuasive essay [40]. Furthermore, Hierarchical Attention Network (HAN) with XGBoost was utilized to similar task and indicated to be a promising method for hierarchical data [41].…”

Section: Argument Analysismentioning

confidence: 99%

Argument annotation and analysis using deep learning with attention mechanism in Bahasa Indonesia

et al. 2020

Self Cite

View full text Add to dashboard Cite

Argumentation mining is a research field which focuses on sentences in type of argumentation. Argumentative sentences are often used in daily communication and have important role in each decision or conclusion making process. The research objective is to do observation in deep learning utilization combined with attention mechanism for argument annotation and analysis. Argument annotation is argument component classification from certain discourse to several classes. Classes include major claim, claim, premise and non-argumentative. Argument analysis points to argumentation characteristics and validity which are arranged into one topic. One of the analysis is about how to assess whether an established argument is categorized as sufficient or not. Dataset used for argument annotation and analysis is 402 persuasive essays. This data is translated into Bahasa Indonesia (mother tongue of Indonesia) to give overview about how it works with specific language other than English. Several deep learning models such as CNN (Convolutional Neural Network), LSTM (Long Short-Term Memory), and GRU (Gated Recurrent Unit) are utilized for argument annotation and analysis while HAN (Hierarchical Attention Network) is utilized only for argument analysis. Attention mechanism is combined with the model as weighted access setter for a better performance. From the whole experiments, combination of deep learning and attention mechanism for argument annotation and analysis arrives in a better result compared with previous research.

show abstract

“…Na literatura são encontrados alguns trabalhos para a mineração de argumentos [Desilia et al 2017;Gema et al 2017…”

Section: Trabalhos Relacionadosunclassified

Processamento do Discurso em Textos Dissertativos-Argumentativos: Uma Abordagem Baseada em Mineração de Argumentos e Aprendizado Supervisionado de Máquina

Nau¹,

Dazzi²,

Filho³

et al. 2020

Anais Do Seminário Integrado De Software E Hardware (SEMISH 2020)

View full text Add to dashboard Cite

A identificação da estrutura da argumentação é útil para analisar o discurso e o raciocínio lógico do aluno. O processo de entendimento do discurso abrange a compreensão da morfologia, da sintaxe e da semântica. Neste trabalho, o processamento do discurso segue uma metodologia composta das seguintes fases: (i) criação de um corpus para treinamento e validação da aprendizagem; (ii) implementação da abordagem para mineração de argumentos; e (iii) aplicação de técnicas de aprendizagem de máquina. Os resultados alcançados ficaram entre a média encontrada na literatura para trabalhos realizados na língua inglesa. O destaque foi a identificação da proposta de intervenção, com um valor-F de 0,744.

show abstract

It Takes Two To Tango: Modification of Siamese Long Short Term Memory Network with Attention Mechanism in Recognizing Argumentative Relations in Persuasive Essay

Cited by 10 publications

References 8 publications

Unlocking Transfer Learning in Argumentation Mining: A Domain-Independent Modelling Approach

Unlocking Transfer Learning in Argumentation Mining: A Domain-Independent Modelling Approach

Argument annotation and analysis using deep learning with attention mechanism in Bahasa Indonesia

Processamento do Discurso em Textos Dissertativos-Argumentativos: Uma Abordagem Baseada em Mineração de Argumentos e Aprendizado Supervisionado de Máquina

Contact Info

Product

Resources

About