Extraction of Online Discussion Structures for Automated Facilitation Agent

Yamaguchi, Naoko; Nishida, Toshirou; Moustafa, Ahmed; Shibata, Daichi; Yoshino, Kai; Hiraishi, Kentaro; Ito, Takayuki

doi:10.1007/978-3-030-39878-1_14

Cited by 9 publications

(18 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To detect the types of the posts, the agent needs to classify the text according to the IBIS types. To this end, we implemented the data extraction module using a Bidirectional Long Short-Term Memory (BiLSTM) classifier (Suzuki et al 2019;Lample et al 2016). The module captures the sentences and their IBIS word constituents (issues, ideas, pros, and cons).…”

Section: Automated Facilitation Agentmentioning

confidence: 99%

See 1 more Smart Citation

An Agent that Facilitates Crowd Discussion

Ito

Hadfi

2021

Group Decis Negot

Self Cite

View full text Add to dashboard Cite

Online discussion platforms are perceived as the next-generation method of citizen involvement. Such platforms can collect, integrate, and synthesize opinions to achieve social good. Crowd-scale platforms are being developed and deployed in social experiments that involve citizens and local governments. In such platforms, human facilitation is often used to preserve the quality of the discussions. Human facilitators often face difficulties when the discussions grow in size. In this paper, we present “D-agree,” a crowd-scale discussion support system based on an automated facilitation agent. The agent extracts discussion structures from online discussions, analyzes them, and posts facilitation messages. We conducted small- and large-scale social experiments in Japan to assess the social impact of the platform. The results showcase the success of our automated facilitation agents in gathering valuable opinions from citizens. In addition, our experiments highlight the effect of an automated facilitation agent on online discussions. In particular, we find that combining the agent facilitator with human facilitators leads to higher user satisfaction.

show abstract

Section: Automated Facilitation Agentmentioning

confidence: 99%

“…Adopting this mapping in the IBIS model provides a richer data model for argumentative discussions. More details on our implementation of the extraction method can be found in a previous study (Suzuki et al 2019).…”

Section: Automated Facilitation Agentmentioning

confidence: 99%

An Agent that Facilitates Crowd Discussion

Ito

Hadfi

2021

Group Decis Negot

Self Cite

View full text Add to dashboard Cite

show abstract

“…Argument structure extraction, as a classic task in argumentation mining which has attracted much attention [1]. For instance, Suzuki et al propose an approach that includes two steps which are node extraction and link extraction [5]. Specifically, they employ bidirectional long short-term memory (Bi-LSTM) for both of node extraction and link extraction tasks.…”

Section: Related Workmentioning

confidence: 99%

“…F or i n s tance, largescale online discussions on the Web such as D-Agree [3] [4] and Slack usually require to process argument structure extraction for efficient d i s cussion. To c onstruct a t ree-like discussion structure, argument structure extraction majorly has two types of tasks : node classification and link prediction [5] [ 6]. In node classification tasks, each argument is as a node and the goal is to accurately classify a label such as topic, idea and issue for each node.…”

Section: Introductionmentioning

confidence: 99%

Graph Convolutional Networks for Link Prediction in Argument Structure Extraction

Ding

Ito²

2023

IIAI Letters on Informatics and Interdisciplinary Research

View full text Add to dashboard Cite

Argument structure extraction is a fundamental problem in argumentation mining which aims at automatically constructing arguments from unstructured textual documents. Re-garding each argument as a node, then predicting relationships of those nodes is a key task for argument structure extraction, which is called link prediction. Graph neural networks (GNN), as an efficient approach to cope with graphical data, have shown good performances on link prediction tasks. However, they usually focus on predicting whether parts of links really exist in one argument graph, rather than predicting existence of the links between any two nodes, though it is more practical in real applications of argument mining. Thus, in this paper our goal is to predict links between any two nodes given only nodes information. This task is more difficult than the traditional link prediction tasks since the prediction scale exponentially increases with node number. We propose a GNN based model for link predic-tion which outputs the embedding of all the nodes. Then, the probability of link existence between any two nodes is calculated by their embedding results. Finally, we use a dataset of essay comments to perform evaluations and the results confirm the effectiveness of our proposed method.

show abstract

“…We then consider a sentence as the type of the node that acquires the highest probability amongst four possible labels (issues, ideas, pros, and cons.) More details on the algorithmic and implementation details are found in Suzuki et al (2019)…”

Section: Natural Language Processing Enginementioning

confidence: 99%

Argumentative Conversational Agents for Online Discussions

Hadfi

Haqbeen

Sahab

et al. 2021

J. Syst. Sci. Syst. Eng.

Self Cite

View full text Add to dashboard Cite

Artificial Intelligence is revolutionising our communication practices and the ways in which we interact with each other. This revolution does not only impact how we communicate, but it affects the nature of the partners with whom we communicate. Online discussion platforms now allow humans to communicate with artificial agents in the form of socialbots. Such agents have the potential to moderate online discussions and even manipulate and alter public opinions. In this paper, we propose to study this phenomenon using a constructed large-scale agent platform. At the heart of the platform lies an artificial agent that can moderate online discussions using argumentative messages. We investigate the influence of the agent on the evolution of an online debate involving human participants. The agent will dynamically react to their messages by moderating, supporting, or attacking their stances. We conducted two experiments to evaluate the platform while looking at the effects of the conversational agent. The first experiment is a large-scale discussion with 1076 citizens from Afghanistan discussing urban policy-making in the city of Kabul. The goal of the experiment was to increase the citizen involvement in implementing Sustainable Development Goals. The second experiment is a small-scale debate between a group of 16 students about globalisation and taxation in Myanmar. In the first experiment, we found that the agent improved the responsiveness of the participants and increased the number of identified ideas and issues. In the second experiment, we found that the agent polarised the debate by reinforcing the initial stances of the participant.

show abstract

Extraction of Online Discussion Structures for Automated Facilitation Agent

Cited by 9 publications

References 17 publications

An Agent that Facilitates Crowd Discussion

An Agent that Facilitates Crowd Discussion

Graph Convolutional Networks for Link Prediction in Argument Structure Extraction

Argumentative Conversational Agents for Online Discussions

Contact Info

Product

Resources

About