User:Jawad Haqbeen/sandbox D-Agree

Stub version
D-Agree is an artificial intelligence-facilitated online debate platform. D-Agree is 'crowd sourcing focused', software platform for public utilities supported by the Nagoya Institute of Technology and Kyoto University.

Working
D-Agree supports large-scale online human to human discourse with a focus to facilitate stakeholder's information-centric participation to reach consensus about urban planning and development decisions.

D-Agree can create abstract of the discussion from the human point of views posted on the online D-Agree platform, the structure creates abstract based on IBIS (issues, ideas, pros, and cons), and also can process data from other social media platforms too.

Deployment in Afghanistan
In a 2022 report titled 'The Transition of Asian and Pacific Cities to a Sustainable Future: Accelerating Action for Sustainable Urbanization' released by United Nations Economic and Social Commission for Asia and the Pacific took note of implementation of D-Agree - for urban planning in Afghanistan as an example of  innovative  approaches and progress in the region. According to The Asahi Shimbun D-Agree was found to be specially helpful for online participation of Afghan women due to real life cultural constraints on them restricting their participation in their otherwise male dominated society.

Detail version
D-Agree is an Artificial intelligence-facilitated online debate platform. D-Agree developed at Nagoya Institute of Technology as an extension of COLLAGREE, and further implementations were developed at Kyoto University, both Japanese public research universities, and supported by the Japan Science and Technology Agency under science funding of research projects. COLLAGREE names comes from COLLective AGREEment, and D-Agree name comes from the Discussion/Diversity/Democracy and -Agreement directories often given to cause of discussion. In a 2022 report titled 'The Transition of Asian and Pacific Cities to a Sustainable Future: Accelerating Action for Sustainable Urbanization' released by United Nations Economic and Social Commission for Asia and the Pacific took note of implementation of D-Agree - for urban planning in Afghanistan as an example of  innovative  approaches and progress in the region.

Membership was initially limited to Nagoya Institute of Technology students and Nagoya city, gradually expanding and to other Japanese cities such as Kasugai, Shima, The Intellectual property of D-Agree is owned by Kyoto University. . In the 2022 edition of QS World University Rankings, Kyoto University was ranked 36th in the world, placing it as the 15th-best Asian university. Furthermore, D-Agree has been used in other countries such as Afghanistan.

History
The first version, COLLAGREE, and initial version of D-Agree was supported under Japan Science and Technology Agency CREST project titled 'Large-scale Consensus Support Systems based on Agent Technologies' for a period of five years [Oct, 2015 - Mar, 2021]. The further implementations of D-Agree will be supported under Japan Science and Technology Agency CREST project titled 'Hyperdemocracy: Implementation of Large-scale Consensus Making Platform based on Social Multiagent Systems' for a period of another five years [Dec, 2020 - Mar, 2026]. COLLAGREE were deployed in Nagoya city. D-Agree were deployed in Nagoya and other Japanese cities than Nagoya such as Kasugai and Shima. In 2019, D-Agree has been used in other countries such as Afghanistan. D-Agree was officially adopted by Kabul city as digital platform to facilitate town meeting to support participatory urban planning in Afghanistan.

Environment
It can be accessed from devices with Internet connectivity, such as personal computers, tablets and smartphones. After registering, users can create a profile revealing information about themselves. They can post text and photos on discussion which are created by administrators. Users can also see their ranking, can agree and disagree with each other with on D-Agree and receive notifications on the activities of their activities they follow. Currently, D-Agree operates in two natural languages, (1) English language, and (2) Japanese language. d-agree.com is platform URL where service provided and that the domain d-agree.com in use. D-Agree employ conversational agent as facilitator which is a computer program to facilitate and extracts a discussion's structure. The facilitation and extraction are based on Issue-based information system, which automatically facilitating and classifying all submitted opinion at real time.

The facilitation performed by a conversational agent that supported to performs goal-driven actions according to predefined plans and policies to guide the discussion towards desired outcomes. The agent interacts directly with the users throughout the D-Agree's website, or external services such as Slack. The agent as facilitator generates replies based on set of rules that dynamically combine dialogue utterances depending on the expected behaviour of the agent to mediate the submitted posts of user based on IBIS. On the other hand, IBIS Structure (Issue, Idea, pros and cons) are extracted with Deep Learning technique called bidirectional Long short-term memory(Bi-LSTM). There are two extraction; (1) link, (2) node. Extracting links means extracting the relationships amongst these nodes which constructed a tree-like discussion structure that consists of the nodes and their relationships in IBIS and employed Bi-LSTM for both of node extraction and link extraction.

Deployment in Afghanistan
The Asahi Shimbun, CNET

D-Agree has been used in Afghanistan since 2019. In a 2022 report released by United Nations Economic and Social Commission for Asia and the Pacific took note of implementation of D-Agree - for urban planning in Afghanistan as an example of  innovative  approaches and progress in the region, which also cited by a paper titled 'Regional Commissions report on the progress on the Implementation of the New Urban Agenda (2019-2022)' released by United Nations took note of implementation of D-Agree. D-Agree has been used to facilitate discussion during the Fall of Kabul (2021) and the news has been covered by major news agencies such as The Asahi Shimbun and CNET Japan.

Draft
D-Agree is a crowd discussion support system based on Artificial intelligence facilitation developed at the Nagoya Institute of Technology and Kyoto University. D-Agree offers many discussion support features, including gathering, facilitating, extracting, and visualizing real-time discussion summaries organized by on-going town-meeting and group discussions.

History
D-Agree was founded in 2018 as the extended version of the former COLLAGREE system. COLLAGREE was a discussion support system used for helping facilitating human facilitator to lead discussion in town planning. The name of the former system combines the letters "Coll" in collective and "Agree" in agreement, and the name of the D-Agree combines the letters "D" in discussion and "A" in agreement.

Both the former and current version is developed under first JST CREST national research grant, called "Innovating Agent-based Large-scale consensus Support System" provided by the funding agency called Japan Science and Technology Agency (JST) The project was for five years (Oct/2015-Mar/2021). The intellectual property (IP) of platform owned by Kyoto University.

COLLAGREE has been used in collaboration with local governments like Japan and D-Agree has been used in collaboration with local governments like Afghanistan for social good.

D-Agree first played a role in the Kabul city town meeting in November 2019 and played a role in planning process in January 2020.

The only difference of D-Agree with former system was server management and facilitation type. Where in COLLAGREE the server was hosted on local computer and D-Agree it was on cloud. Most importantly, the facilitation was done by humans in COLLAGREE while in D-Agree it is done by AI where software entities lead the facilitation.

Features
It is a platform that enables active exchange of opinions online by utilizing AI. It is often used for the purposes of "discussing and drawing conclusions" and "gathering opinions and conducting large-scale surveys". Online users can post text in a discussion space moderated by an AI. The main goal of the AI is to drive the discussions towards group consensus. The labeling of the posted text is performed according to the Issue-based information system (IBIS) model. The discussion summarization is also done in real-time. D-Agree has an incentive mechanism where users can also see their ranking. The users can also agree and disagree with each other and receive notifications about the activities they follow.

System Architecture
D-Agree is composed of two modules: front-end and back-end architectures. The former is the system user interface (UI), which interacts with users through discussion websites or other social platform messaging services. The back-end hosts four agent modules: The top part of the picture in the Figure shows the front-end, and the bottom part shows the back-end architectures of D-Agree system.
 * 1) A conversational agent
 * 2) A proactive agent
 * 3) An NLP engine
 * 4) An argumentation engine

Real-world Experiment in collaboration with Governments
Firstly, D-Agree has been used in Japan, but gradually the system expanded to other countries like municipal government of Kabul city Afghanistan.

Protect Users with Taliban take control of Afghanistan
D-Agree announced on 15 August 2021 that it will make an effort to protect all community users with Taliban take control of Afghanistan by anonymizing all users ID.

Media publication
Demonstration experiment with Nagoya City using COLLAGREE was published in city official page. Demonstration experiment with Kasugai City using D-Agree was published in "Chunichi Shimbun". An article about the real-experiment in Afghanistan using D-Agree was published in the Asahi Shimbun GLOBE +, a web media that conveys the present of the world. An interview article with representative Kuwahara was posted on the startup & tech site "Techable" discussing about D-Agree activities.

A demonstration video posted on Kabul city municipality on how to login on D-Agree system for idea contest project. An article about the efforts related first internet based crowdsourcing idea contest was published in Kabul city homepage

D-Agree's Next Version (X-Agree)
The future version will be towards more sophisticated AI-based facilitation to lead discussion at scale.