Argument technology

Argument technology is a sub-field of collective intelligence and artificial intelligence that focuses on applying computational techniques to the creation, identification, analysis, navigation, evaluation and visualisation of arguments and debates.

In the 1980s and 1990s, philosophical theories of arguments in general, and argumentation theory in particular, were leveraged to handle key computational challenges, such as modeling non-monotonic and defeasible reasoning and designing robust coordination protocols for multi-agent systems.[1] At the same time, mechanisms for computing semantics of Argumentation frameworks were introduced as a way of providing a calculus of opposition for computing what it is reasonable to believe in the context of conflicting arguments.[2]

With these foundations in place, the area was kick-started by a workshop held in the Scottish Highlands in 2000, the result of which was a book coauthored by philosophers of argument, rhetoricians, legal scholars and AI researchers.[3] Since then, the area has been supported by various dedicated events such as the International Workshop on Computational Models of Natural Argument (CMNA)[4] which has run annually since 2001; the International Workshop on Argument in Multi Agent Systems (ArgMAS) annually since 2004; the Workshop on Argument Mining,[5] annually since 2014, and the Conference on Computational Models of Argument (COMMA),[6] biennially since 2006. Since 2010, the field has also had its own journal, Argument & Computation, which was published by Taylor & Francis until 2016[7] and since then by IOS Press.[8]

One of the challenges that argument technology faced was a lack of standardisation in the representation and underlying conception of argument in machine readable terms. Many different software tools for manual argument analysis, in particular, developed idiosyncratic and ad hoc ways of representing arguments which reflected differing underlying ways of conceiving of argumentative structure.[9] This lack of standardisation also meant that there was no interchange between tools or between research projects, and little re-use of data resources that were often expensive to create. To tackle this problem, the Argument Interchange Format[10] set out to establish a common standard that captured the minimal common features of argumentation which could then be extended in different settings.

Since about 2018, argument technology has been growing rapidly, with, for example, IBM's Grand Challenge, Project Debater, results for which were published in Nature in March 2021;[11] German research funder, DFG's nationwide research programme on Robust Argumentation Machines, RATIO,[12] begun in 2019; and UK nationwide deployment of The Evidence Toolkit by the BBC in 2019.[13] A 2021 video narrated by Stephen Fry provides a summary of the societal motivations for work in argument technology.[14]

Argument technology has applications in a variety of domains, including education, healthcare, policy making, political science, intelligence analysis and risk management and has a variety of sub-fields, methodologies and technologies.[15]

Technologies

Argument assistant
See also: Proof assistant and Computer-assisted proof

An argument assistant is a software tool which support users when writing arguments. Argument assistants can help users compose content, review content from one other, including in dialogical contexts. In addition to Web services, such functionalities can be provided through the plugin architectures of word processor software or those of Web browsers. Internet forums, for instance, can be greatly enhanced by such software tools and services.

Argument blogging
See also: Argument Web

ArguBlogging is software which allows its users to select portions of hypertext on webpages in their Web browsers and to agree or disagree with the selected content, posting their arguments to their blogs with linked argument data.[16] It is implemented as a bookmarklet, adding functionality to Web browsers and interoperating with blogging platforms such as Blogger and Tumblr.[16]

Argument mapping
Main article: Argument map
Kialo debate tree schema with an example path through it: all Con-argument boxes and some Pros were emptied to illustrate an example path.[17]

Argument maps are visual, diagrammatic representations of arguments. Such visual diagrams facilitate diagrammatic reasoning and promote one's ability to grasp and to make sense of information rapidly and readily. Argument maps can provide structured, semi-formal frameworks for representing arguments using interactive visual language. One avenue of research and development is the design of online platforms to leverage collective intelligence to populate such maps and to integrate data, optimize and assess arguments.

Argument mining
Main article: Argument mining
See also: Argumentation theory § Artificial intelligence

Argument mining, or argumentation mining, is a research area within the natural language processing field. The goal of argument mining is the automatic extraction and identification of argumentative structures from natural language text with the aid of computer programs.

Main article: Search engine (computing)

An argument search engine is a search engine that is given a topic as a user query and returns a list of arguments for and against the topic[18][19] or about that topic.[20] Such engines could be used to support informed decision-making or to help debaters prepare for debates.

Automated argumentative essay scoring
Main article: Automated essay scoring

The goal of automated argumentative essay scoring systems is to assist students in improving their writing skills by measuring the quality of their argumentative content.[21][22]

Debate technology
Structured debates from platforms like Kialo could be used for "artificial deliberative agents" (ADAs) or computational reasoning.[23][24]
Example of an ADA contributing missing information to a debate via crawled Kialo data and selected based on the prior conversation and crawled argument weight ratings.[24]

Debate technology focuses on human-machine interaction and in particular providing systems that support, monitor and engage in debate. One of the most high-profile examples of debating technology is IBM's Project Debater[11] which combines scripted communication with very large-scale processing of news articles to identify and construct arguments on the fly in a competitive debating setting. Debating technology also encompasses tools aimed at providing insight into debates, typically using techniques from data science. These analytics have been developed in both academic[25] and commercial[26] settings.

Decision support system
Main articles: Decision support system and Intelligent decision support system

Argument technology can reduce both individual and group biases and facilitate more accurate decisions. Argument-based decision support systems do so by helping users to distinguish between claims and the evidence supporting them, and express their confidence in and evaluate the strength of evidence of competing claims.[27] They have been used to improve predictions of housing market trends,[27] risk analysis,[28] ethical and legal decision making.

Ethical decision support system
See also: Moral reasoning

An ethical decision support system is a decision support system which supports users in moral reasoning and decision-making.[29][30]

Main articles: Legal expert system and Applications of artificial intelligence to legal informatics

A legal decision support system is a decision support system which supports users in legal reasoning and decision-making.

Explainable artificial intelligence
Main articles: Explainable artificial intelligence, Argument § Explanations, and Free and open source software

An explainable or transparent artificial intelligence system is an artificial intelligence system whose actions can be easily understood by humans.

Intelligent tutoring system
Main article: Intelligent tutoring system

An intelligent tutoring system is a computer system that aims to provide immediate and customized instruction or feedback to learners, usually without requiring intervention from a human teacher. The intersection of argument technology and intelligent tutoring systems includes computer systems which aim to provide instruction in: critical thinking, argumentation,[31] ethics,[32] law,[33] mathematics,[34] and philosophy.

Main article: Legal expert system

A legal expert system is a domain-specific expert system that uses artificial intelligence to emulate the decision-making abilities of a human expert in the field of law.

Machine ethics
Main article: Machine ethics
See also: Moral reasoning

Machine ethics is a part of the ethics of artificial intelligence concerned with the moral behavior of artificially intelligent beings. As humans argue with respect to morality and moral behavior, argument can be envisioned as a component of machine ethics systems and moral reasoning components.

Proof assistant
Main articles: Proof assistant and Computer-assisted proof

In computer science and mathematical logic, a proof assistant or interactive theorem prover is a software tool to assist with the development of formal proofs by human-machine collaboration. This involves some sort of interactive proof editor, or other interface, with which a human can guide the search for proofs, the details of which are stored in, and some steps provided by, a computer.

Ethical considerations
Main article: Machine ethics

Ethical considerations of argument technology include privacy, transparency, societal concerns, and diversity in representation. These factors cut across different levels such as technology, user interface design, user, service context, and society.[35] There is concern about unethical misuse for "generating arguments on controversial topics with specific stances and deploying them on social platforms".[36] Another issue may concern the design of conclusion-making algorithms, such as e.g. enabling such to conclude that certain key data is needed instead of only making lists of best-fit conclusions or enabling the generation of multiple conclusions from the same data based on different argument-assessments or assessment methods.

References
  1. van Eemeren, Frans H.; Garssen, Bart; Krabbe, Erik C. W.; Snoeck Henkemans, A. Francisca; Verheij, Bart; Wagemans, Jean H. M. (2014), van Eemeren, Frans H.; Garssen, Bart; Krabbe, Erik C. W.; Snoeck Henkemans, A. Francisca (eds.), "Argumentation and Artificial Intelligence", Handbook of Argumentation Theory, Dordrecht: Springer Netherlands, pp. 615–675, doi:10.1007/978-90-481-9473-5_11, ISBN 978-90-481-9473-5, retrieved 2021-12-08
  2. Prakken, Henry (2014). "The ASPIC+ framework for structured argumentation: a tutorial". Argument & Computation. 5 (1): 31–62. doi:10.1080/19462166.2013.869766.
  3. Reed, C. & Norman, T.J. (eds) Argumentation Machines. Kluwer, 2003.
  4. "Computational Models of Natural Argument". www.cmna.info.
  5. For example: "Proceedings of the 6th Workshop on Argument Mining – ACL Anthology". www.aclweb.org. Association for Computational Linguistics. August 2019. Retrieved 7 December 2020.
  6. "Computational Models of Argument conference series". www.comma-conf.org.
  7. "Journal of Argument & Computation". www.tandf.co.uk. Archived from the original on 2012-02-21.
  8. "Journal of Argument & Computation". www.iospress.nl. August 2023.
  9. Scheuer, O.; Loll, F.; Pinkwart, N.; McLaren, B.M. (2010). "Computer-supported argumentation: A review of the state of the art". Computer-Supported Collaborative Learning. 5 (1): 43–102. doi:10.1007/s11412-009-9080-x. S2CID 4473082.
  10. Rahwan, Iyad; Reed, Chris (2009), Simari, Guillermo; Rahwan, Iyad (eds.), "The Argument Interchange Format", Argumentation in Artificial Intelligence, Boston, MA: Springer US, pp. 383–402, Bibcode:2009aai..book..383R, doi:10.1007/978-0-387-98197-0_19, hdl:21.11116/0000-0002-CA71-B, ISBN 978-0-387-98197-0, retrieved 2021-12-08
  11. Slonim, Noam; Bilu, Yonatan; Alzate, Carlos; Bar-Haim, Roy; Bogin, Ben; Bonin, Francesca; Choshen, Leshem; Cohen-Karlik, Edo; Dankin, Lena; Edelstein, Lilach; Ein-Dor, Liat (March 2021). "An autonomous debating system". Nature. 591 (7850): 379–384. Bibcode:2021Natur.591..379S. doi:10.1038/s41586-021-03215-w. ISSN 1476-4687. PMID 33731946. S2CID 232305184.
  12. "Priority Programme "Robust Argumentation Machines (RATIO)" (SPP 1999/2)". www.dfg.de. Retrieved 2021-12-08.
  13. Visser, Jacky; Lawrence, John; Reed, Chris (2020-10-22). "Reason-checking fake news". Communications of the ACM. 63 (11): 38–40. doi:10.1145/3397189. ISSN 0001-0782. S2CID 225041563.
  14. AI Exposes A Group Behind Covid, Brexit & Climate, w Stephen Fry., retrieved 2021-12-08
  15. Bex, Floris J.; Grasso, Floriana; Green, Nancy L.; Paglieri, Fabio; Reed, Chris, eds. (2017). Argument Technologies: Theory, Analysis, and Applications. Studies in logic and argumentation. Vol. 68. London: College Publications. ISBN 9781848902183. OCLC 1012498399.
  16. Bex, Floris J.; Snaith, Mark; Lawrence, John; Reed, Chris (March 2014). "ArguBlogging: an application for the argument web" (PDF). Web Semantics: Science, Services and Agents on the World Wide Web. 25: 9–15. doi:10.1016/j.websem.2014.02.002.
  17. Bolton, Eric; Calderwood, Alex; Christensen, Niles; Kafrouni, Jerome; Drori, Iddo (2020). "High quality real-time structured debate generation". arXiv:2012.00209 [cs.CL].
  18. Aharoni, Ehud; et al. (2014). "Claims on demand–an initial demonstration of a system for automatic detection and polarity identification of context dependent claims in massive corpora". Proceedings of COLING 2014: 6–9.
  19. Ajjour, Yamen (2019). "Data acquisition for argument search: the args.me corpus". 42nd German Conference on Artificial Intelligence (KI 2019): 48–59.
  20. Lawrence, John; Bex, Floris; Reed, Chris; Snaith, Mark (September 2012). "AIFdb: Infrastructure for the Argument Web". In Verheij, Bart; Szeider, Stefan; Woltran, Stefan (eds.). Computational Models of Argument. Frontiers in Artificial Intelligence and Applications. Vol. 245. IOS Press. pp. 515–516. doi:10.3233/978-1-61499-111-3-515.
  21. Stab, Christian; Gurevych, Iryna (2014). "Identifying argumentative discourse structures in persuasive essays". Proceedings of EMNLP 2014: 46–56. doi:10.3115/v1/D14-1006. S2CID 71907.
  22. Green, Nancy L. (2013). "Towards automated analysis of student arguments". In Lane, H. Chad; Yacef, Kalina; Mostow, Jack; Pavlik, Philip (eds.). Artificial intelligence in education: 16th international conference, AIED 2013, Memphis, TN, USA, July 9–13, 2013: proceedings. Lecture notes in computer science. Vol. 7926. Berlin; New York: Springer-Verlag. pp. 591–594. doi:10.1007/978-3-642-39112-5_66. ISBN 9783642391118.
  23. Anastasiou, Lucas; De Liddo, Anna (8 May 2021). "Making Sense of Online Discussions: Can Automated Reports help?". Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems. Association for Computing Machinery. pp. 1–7. doi:10.1145/3411763.3451815. ISBN 9781450380959. S2CID 233987842.
  24. Betz, Gregor (2022). "Natural-language multi-agent simulations of argumentative opinion dynamics". Journal of Artificial Societies and Social Simulation. 25: 2. arXiv:2104.06737. doi:10.18564/jasss.4725. S2CID 233231231.
  25. For example: "Argument Analytics". analytics.arg.tech. Retrieved 11 August 2020.
  26. For example: "Kialo". www.kialo.com. Retrieved 11 August 2020.
  27. Introne, Joshua; Iandoli, Luca (2014-08-01). "Improving decision-making performance through argumentation: An argument-based decision support system to compute with evidence". Decision Support Systems. 64: 79–89. doi:10.1016/j.dss.2014.04.005. ISSN 0167-9236.
  28. Hansson, Sven Ove; Hirsch Hadorn, Gertrude (2018-12-02). "Argument-based decision support for risk analysis". Journal of Risk Research. 21 (12): 1449–1464. doi:10.1080/13669877.2017.1313767. ISSN 1366-9877. S2CID 125354345.
  29. Mancherjee, Kevin; Sodan, Angela C. (September 2004). "Can computer tools support ethical decision making?". ACM SIGCAS Computers and Society. 34 (2): 1. CiteSeerX 10.1.1.61.7160. doi:10.1145/1052791.1052792. S2CID 5503640.
  30. Mathieson, Kieran (December 2007). "Towards a design science of ethical decision support". Journal of Business Ethics. 76 (3): 269–292. doi:10.1007/s10551-006-9281-4. JSTOR 25075516. S2CID 143999917.
  31. Loll, Frank; Pinkwart, Niels; Scheuer, Oliver; McLaren, Bruce M. (July 2009). "Towards a flexible intelligent tutoring system for argumentation". 2009 Ninth IEEE International Conference on Advanced Learning Technologies, 15–17 July 2009. IEEE. pp. 647–648. CiteSeerX 10.1.1.329.6574. doi:10.1109/ICALT.2009.138. S2CID 1026429.
  32. Goldin, Ilya M.; Ashley, Kevin D.; Pinkus, Rosa L. (May 2001). "Introducing PETE: computer support for teaching ethics". ICAIL '01: Proceedings of the 8th International Conference on Artificial Intelligence and Law. New York: Association for Computing Machinery. pp. 94–98. CiteSeerX 10.1.1.19.6676. doi:10.1145/383535.383546. ISBN 978-1-58113-368-4. S2CID 979724.
  33. Ashley, Kevin D.; Aleven, Vincent (May 1991). "Toward an intelligent tutoring system for teaching law students to argue with cases". ICAIL '91: Proceedings of the 3rd International Conference on Artificial Intelligence and Law. New York: Association for Computing Machinery. pp. 42–52. doi:10.1145/112646.112651. ISBN 978-0-89791-399-7. S2CID 6814858.
  34. Ritter, Steven; Anderson, John R.; Koedinger, Kenneth R.; Corbett, Albert (April 2007). "Cognitive tutor: applied research in mathematics education" (PDF). Psychonomic Bulletin & Review. 14 (2): 249–255. CiteSeerX 10.1.1.158.4283. doi:10.3758/bf03194060. PMID 17694909. S2CID 9547046.
  35. Brüggemeier, Birgit; Lalone, Philip (January 2022). "Perceptions and reactions to conversational privacy initiated by a conversational user interface". Computer Speech & Language. 71: 101269. doi:10.1016/j.csl.2021.101269. ISSN 0885-2308.
  36. Al Khatib, Khalid; Trautner, Lukas; Wachsmuth, Henning; Hou, Yufang; Stein, Benno (August 2021). "Employing argumentation knowledge graphs for neural argument generation" (PDF). Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). Association for Computational Linguistics: 4744–4754. doi:10.18653/v1/2021.acl-long.366. S2CID 236460348.
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.