I’m co-author of a paper in a post-workshop proceedings of the Workshop on Computational Models of Natural Argument in 2010 and 2011.
Working on the Argument Pipeline: Through Flow Issues between Natural Language Argument, Instantiated Arguments, and Argumentation Frameworks
Adam Wyner, Tom van Engers, and Anthony Hunter
Abstract
In many domains of public discourse such as arguments about public policy, there is an abundance of knowledge to store, query, and reason with. To use this knowledge, we must address two key general problems: first, the problem of the knowledge acquisition bottleneck between forms in which the knowledge is usually expressed, e.g. natural language, and forms which can be automatically processed; second, reasoning with the uncertainties and inconsistencies of the knowledge. Given such complexities, it is labour and knowledge intensive to conduct policy consultations, where participants contribute statements to the policy discourse. Yet, from such a consultation, we want to derive policy positions, where each position is a set of consistent statements, but where positions may be mutually inconsistent. To address these problems and support policy-making consultations, we consider recent automated techniques in natural language processing, instantiating arguments, and reasoning with the arguments in argumentation frameworks. We discuss application and “bridge” issues between these techniques, outlining a pipeline of technologies whereby: expressions in a controlled natural language are parsed and translated into a logic (a literals and rules knowledge base), from which we generate instantiated arguments and their relationships using a logic-based formalism (an argument knowledge base), which is then input to an implemented argumentation framework that calculates extensions of arguments (an argument extensions knowledge base), and finally, we extract consistent sets of expressions (policy positions). The paper reports progress towards reasoning with web-based, distributed, collaborative, incomplete, and inconsistent knowledge bases expressed in natural language.
Bibtex
@INPROCEEDINGS{WynerVanEngersHunterCMNAPOST2013,
author = {Adam Wyner and Tom van Engers and Anthony Hunter},
title = {Working on the Argument Pipeline: Through Flow Issues between Natural
Language Argument, Instantiated Arguments, and Argumentation Frameworks},
booktitle = {Proceedings of the Workshop on Computational Models of Natural Argument},
year = {2013},
editor = {??},
pages = {??-??},
note = {To appear}
}
Shortlink to this page.
By Adam Wyner
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Category: controlled natural language
Presentations at CMNA 2012 and RuleML 2012
I have gave a talk about my paper at the ECAI workshop on Computational Models of Natural Argumentation 2012 and also presented an invited talk at the RuleML 2012 conference. The PDFs of these talks are below.
Questions, arguments, and natural language semantics
Translating Rules in Natural Language to RuleML
Shortlink to this page.
By Adam Wyner
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
EXTENDED CFP – Workshop on Semantic Processing of Legal Texts (SPLeT 2012)
In conjunction with
Language Resources and Evaluation Conference 2012 (LREC 2012)
27 May, 2012
Istanbul, Turkey
REVISED SUBMISSION DEADLINE FOR WORKSHOP: 19 February 2012
Context
The legal domain represents a primary candidate for web-based information distribution, exchange and management, as testified by the numerous e-government, e-justice and e-democracy initiatives worldwide. The last few years have seen a growing body of research and practice in the field of Artificial Intelligence and Law which addresses a range of topics: automated legal reasoning and argumentation, semantic and cross-language legal information retrieval, document classification, legal drafting, legal knowledge discovery and extraction, as well as the construction of legal ontologies and their application to the law domain. In this context, it is of paramount importance to use Natural Language Processing techniques and tools that automate and facilitate the process of knowledge extraction from legal texts.
Since 2008, the SPLeT workshops have been a venue where researchers from the Computational Linguistics and Artificial Intelligence and Law communities meet, exchange information, compare perspectives, and share experiences and concerns on the topic of legal knowledge extraction and management, with particular emphasis on the semantic processing of legal texts. Within the Artificial Intelligence and Law community, there have also been a number of dedicated workshops and tutorials specifically focussing on different aspects of semantic processing of legal texts at conferences such as JURIX-2008, ICAIL-2009, ICAIL-2011, as well as in the International Summer School “Managing Legal Resources in the Semantic Web” (2007, 2008, 2009, 2010, 2011).
To continue this momentum and to advance research, a 4th Workshop on “Semantic Processing of Legal Texts” is being organized at the LREC-2012 conference to bring to the attention of the broader LR/HLT (Language Resources/Human Language Technology) community the specific technical challenges posed by the semantic processing of legal texts and also share with the community the motivations and objectives which make it of interest to researchers in legal informatics. The outcome of these interactions are expected to advance research and applications and foster interdisciplinary collaboration within the legal domain.
New to this edition of the workshop are two sub-events (described below) to provide common and consistent task definitions, datasets, and evaluation for legal-IE systems along with a forum for the presentation of varying but focused efforts on their development.
The main goals of the workshop and associated events are to provide an overview of the state-of-the-art in legal knowledge extraction and management, to explore new research and development directions and emerging trends, and to exchange information regarding legal language resources and human language technologies and their applications.
Sub-events
Dependency Parsing
The first sub-event will be a shared task specifically focusing on dependency parsing of legal texts: although this is not a domain-specific task, it is a task which creates the prerequisites for advanced IE applications operating on legal texts, which can benefit from reliable preprocessing tools. For this year our aim is to create the prerequisites for more advanced domain-specific tasks (e.g. event extraction) to be organized in future SPLeT editions. We strongly believe that this could be a way to attract the attention of the LR/HLT community to the specific challenges posed by the analysis of this type of texts and to have a clearer idea of the current state of the art. The languages dealt with will be Italian and English. A specific Call for Participation for the shared task is available in a dedicated page.
Semantic Annotation
The second sub-event will be an online, manual, collaborative, semantic annotation exercise, the results of which will be presented and discussed at the workshop. The goals of the exercise are: (1) to gain insight on and work towards the creation of a gold standard corpus of legal documents in a cohesive domain; and (2) to test the feasibility of the exercise and to get feedback on its annotation structure and workflow. The corpus to be annotated will be a selection of documents drawn from EU and US legislation, regulation, and case law in a particular domain (e.g. consumer or environmental protection). For this exercise, the language will be English. A specific Call for Participation for this annotation exercise is available in a dedicated page.
Areas of Interest
The workshop will focus on the topics of the automatic extraction of information from legal texts and the structural organisation of the extracted knowledge. Particular emphasis will be given to the crucial role of language resources and human language technologies.
Papers are invited on, but not limited to, the following topics:
LREC Conference Information (Accommodation, Travel, Registration)
Language Resources and Evaluation Conference 2012 (LREC 2012)
Workshop Schedule – TBA
Workshop Registration and Location – TBA
Webpage URLs
Important Dates:
Author Guidelines:
Submissions are solicited from researchers working on all aspects of semantic processing of legal texts. Authors are invited to submit papers describing original completed work, work in progress, interesting problems, case studies or research trends related to one or more of the topics of interest listed above. The final version of the accepted papers will be published in the Workshop Proceedings.
Short or full papers can be submitted. Short papers are expected to present new ideas or new visions that may influence the direction of future research, yet they may be less mature than full papers. While an exhaustive evaluation of the proposed ideas is not necessary, insight and in-depth understanding of the issues is expected. Full papers should be more well developed and evaluated. Short papers will be reviewed the same way as full papers by the Program Committee and will be published in the Workshop Proceedings.
Full paper submissions should not exceed 10 pages, short papers 6 pages. See the style guidelines and files on the LREC site:
Authors’ Kit and Templates
Submit papers to:
Submission for the workshop uses the START submission system at:
https://www.softconf.com/lrec2012/LegalTexts2012/
Note that when submitting a paper through the START page, authors will be asked to provide essential information about resources (in a broad sense, i.e. also technologies, standards, evaluation kits, etc.) that have been used for the work described in the paper or are a new result of your research. For further information on this new initiative, please refer to:
http://www.lrec-conf.org/lrec2012/?LRE-Map-2012
Publication:
After the workshop a number of selected, revised, peer-reviewed articles will be published in a Special Issue on Semantic Processing of Legal Texts of the AI and Law Journal (Springer).
Contact Information:
Address any queries regarding the workshop to:
lrec_legalWS@ilc.cnr.it
Program Committee Co-Chairs:
Enrico Francesconi (National Research Center, Italy)
Simonetta Montemagni (National Research Center, Italy)
Wim Peters (University of Sheffield, UK)
Adam Wyner (University of Liverpool, UK)
Program Committee (Preliminary):
Kevin Ashley (University of Pittsburgh, USA)
Johan Bos (University of Rome, Italy)
Daniele Bourcier (Humboldt Universitat, Germany)
Pompeu Casanovas (Universitat Autonoma de Barcelona, Spain)
Jack Conrad (Thomson Reuters, USA)
Matthias Grabmair (University of Pittsburgh, USA)
Antonio Lazari (Scuola Superiore S.Anna, Italy)
Leonardo Lesmo (Universita di Torino, Italy)
Marie-Francine Moens (Katholieke Universiteit Leuven, Belgium)
Thorne McCarty (Rutgers University, USA)
Raquel Mochales Palau (Catholic University of Leuven, Belgium)
Paulo Quaresma (Universidade de Evora, Portugal)
Tony Russell-Rose (UXLabs, UK)
Erich Schweighofer (Universitat Wien, Austria)
Rolf Schwitter (Macquarie University, Australia)
Manfred Stede (University of Potsdam, Germany)
Daniela Tiscornia (National Research Council, Italy)
Tom van Engers (University of Amsterdam, Netherlands)
Giulia Venturi (Scuola Superiore S.Anna, Italy)
Vern R. Walker (Hofstra University, USA)
Radboud Winkels (University of Amsterdam, Netherlands)
By Adam Wyner
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Workshop Applying Human Language Technology to the Law
A workshop at
ICAIL 2011: The Thirteenth International Conference on Artificial Intelligence and Law
Applying Human Language Technology to the Law (AHLTL 2011)
June 10, 2011
University of Pittsburgh School of Law
Overview:
Over the last decade there have been dramatic improvements in the effectiveness and accuracy of Human Language Technology (HLT), accompanied by a significant expansion of the HLT community itself. Over the same period, there have been widespread developments in web-based distribution and processing of legal textual information, e.g. cases, legislation, citizen information sources, etc. More recently, a growing body of research and practice has addressed a range of topics common to both the HLT and Artificial Intelligence and Law communities, including automated legal reasoning and argumentation, semantic information retrieval, cross and multi-lingual information retrieval, document classification, logical representations of legal language, dialogue systems, legal drafting, legal knowledge discovery and extraction, linguistically based legal ontologies, among others. Central to these shared topics is use of HLT techniques and tools for automating knowledge extraction from legal texts and for processing legal language.
The workshop has several objectives. The first objective is to broaden the research base by introducing HLT researchers to the materials and problems of processing legal language. The second objective is to introduce AI and Law researchers to up-to-date theories, techniques, and tools from HLT, which can be applied to legal language. And the third objective is to deepen the existing research streams. Altogether, the interactions among the researchers are expected to advance research and applications and foster interdisciplinary collaboration within the legal domain.
Context:
Over the last two years, there have been several workshops and tutorials on or relating to processing legal texts and legal language, demonstrating a significant surge of interest. There have been two workshops on Semantic processing of legal texts (SPLeT) held in conjunction with LREC (2008 in Marrakech, Morocco; and 2010 in Malta). At ICAIL 2009, there were two workshops, LOAIT ’09 – the 3rd Workshop on Legal Ontologies and Artificial Intelligence Techniques joint with the 2nd Workshop on Semantic Processing of Legal Texts and NALEA ’09 – Workshop on the Natural Language Engineering of Legal Argumentation: Language, Logic, and Computation. LOAIT ’09 focussed on Legal Knowledge Representation with particular emphasis on the issue of ontology acquisition from legal texts, while NALEA ’09 tackled issues related to legal argumentation. In 2009, the National Science Foundation sponsored a workshop Automated Content Analysis and the Law, which drew participants from computational linguistics and political science. Finally, at the Second Workshop on Controlled Natural Language (CNL 2010), there were several presentations related to legal language.
Intended Audience:
The intended audience would include both current members of the AI & law community who are interested in automated analysis of legal texts and corpora and, in addition, HLT researchers for whom analysis of legal texts would provide an opportunity for development and evaluation of HLT techniques. It is anticipated that participants would come from industry (e.g. The MITRE Corporation, Thomson/Reuters, Endeca, Lexis/Nexis, Oracle), the judiciary in the US and Europe, national organisations (e.g. the US National Institute of Standards and Technology, the US National Science Foundation, European Science Foundation, the UK Office of Public Sector Information), government security agencies, legal professionals, and academic HLT researchers.
Areas of Interest:
The workshop will focus on extraction of information from legal text, representations of legal language (ontologies and semantic translations), and dialogic aspects. While information extraction and retrieval are crucial areas, the workshop emphasises syntactic, semantic, and dialogic aspects of legal information processing.
- Building legal resources: terminologies, ontologies, corpora.
- Ontologies of legal texts, including subareas such as ontology acquisition, ontology customisation, ontology merging, ontology extension, ontology evolution, lexical information, etc.
- Information retrieval and extraction from legal texts.
- Semantic annotation of legal texts.
- Multilingual aspects of legal text semantic processing.
- Legal thesauri mapping.
- Automatic Classification of legal documents.
- Automated parsing and translation of natural language arguments into a logical formalism.
- Linguistically-oriented XML mark up of legal arguments.
- Computational theories of argumentation that are suitable to natural language.
- Controlled language systems for law.
- Name matching and alias detection.
- Dialogue protocols and systems for legal discussion.
Workshop Schedule
- 9:00 Opening remarks
- 9:15 Jack Conrad (invited speaker). The Role of HLT in High-end Search and the Persistent Need for Advanced HLT Technologies
- 10:00 Tommaso Fornaciari and Massimo Poesio. Lexical vs. Surface Features in Deceptive Language Analysis
- 10:30 Nuria Casellas, Joan-Josep Vallbé and Thomas Bruce. Legal Thesauri Reuse. An Experiment with the U.S. Code of Federal Regulations
- 11:00 Break
- 11:15 Meritxell Fernández-Barrera and Pompeu Casanovas. Towards the intelligent processing of non-expert generated content: mapping web 2.0 data with ontologies in the domain of consumer mediation
- 11:45 Emile De Maat and Radboud Winkels. Formal Models of Sentences in Dutch Law
- 12:15 Guido Boella, Llio Humphreys, Leon Van Der Torre and Piercarlo Rossi. Eunomos, a legal document management system based on legislative XML and ontologies (Position paper)
- 12:45 Anna Ronkainen. From Spelling Checkers to Robot Judges? Some Implications of Normativity in Language Technology and AI and Law
- 13:15 Lunch
Workshop Location
To be announced.
Author Guidelines:
- The workshop solicits full papers and position papers. Authors are welcome to submit tentative, incremental, and exploratory studies which examine HLT issues distinctive to the law and legal applications. Papers not accepted as full papers may be accepted as short research abstracts. Submissions will be evaluated by the program committee. For information on submission details (length, format, notion of position paper, etc) see the ICAIL 2011 conference information:
ICAIL CFP
- Submissions should be submitted electronically in PDF to the EasyChair site by the deadline (see important dates below):
AHLTL 2011, an EasyChair site
Publication:
- Selected papers are to be invited to be revised and submitted to a special edition of the AI and Law journal, edited by Adam Wyner and Karl Branting.
The papers from the workshop are available from here.
Webpage:
Important Dates:
- Final version deadline: 23 May 2011
- Workshop date: 10 June 2011
Contact Information:
- Primary contact: Adam Wyner, adam@wyner.info
- Secondary contact: Karl Branting, lbranting@mitre.org
Program Committee Co-Chairs:
- Adam Wyner (University of Liverpool, UK)
- Karl Branting (The MITRE Corporation, USA)
Program Committee:
- Kevin Ashley (University of Pittsburgh, USA)
- Johan Bos (University of Rome, Italy)
- Sherri Condon (The MITRE Corporation, USA)
- Jack Conrad (Thomson Reuters, USA)
- Enrico Francesconi (ITTIG-CNR, Florence, Italy)
- Ben Hachey (Macquarie University, Australia)
- Alessandro Lenci (Università di Pisa, Italy)
- Leonardo Lesmo (Università di Torino, Italy)
- Emile de Maat (University of Amsterdam, Netherlands)
- Thorne McCarty (Rutgers University, USA)
- Marie-Francine Moens (Catholic University of Leuven, Belgium)
- Simonetta Montemagni (ILC-CNR, Italy)
- Raquel Mochales Palau (Catholic University of Leuven, Belgium)
- Craig Pfeifer (The MITRE Corporation, USA)
- Wim Peters (University of Sheffield, United Kingdom)
- Paulo Quaresma (Universidade de Évora, Portugal)
- Mike Rosner (University of Malta, Malta)
- Tony Russell-Rose (Endeca, United Kingdom)
- Erich Schweighofer (Universität Wien, Austria)
- Rolf Schwitter (Macquarie University, Australia)
- Manfred Stede (University of Potsdam, Germany)
- Mihai Surdeanu (Stanford University, USA)
- Daniela Tiscornia (ITTIG-CNR, Italy)
- Radboud Winkels (University of Amsterdam, Netherlands)
- Jonathan Zeleznikow (Victoria University, Australia)
Computational Argumentation on the Web with Natural Language
Over the last four years, I have been working on topics related to computational argumentation on the web using natural language. Some of my publications and previous postings reflect these interests. Along with my colleague Tom van Engers, I prepared two research proposals on this topic, which are here presented as technical reports of our work. These reports are also relevant to the current IMPACT project, which addresses many of the same themes.
There is a short paper (five pages) which outlines key ideas, but has little in the way of discussion or background discussion. There is a long paper (28 pages) which goes into the proposal in much more depth.
Comments and discussion on these documents are very welcome.
By Adam Wyner
Distributed under the Creative Commons
Attribution-Non-Commercial-Share Alike 2.0
Recent Paper Submissions
During my time at the Leibniz Center for Law working on the IMPACT, I and my colleagues Tom van Engers and Kiavash Bahreini prepared and submitted three papers to conferences and workshops. The drafts of the papers are linked below along with the abstracts. Comments welcome.
A Framework for Enriched, Controlled On-line Discussion Forums for e-Government Policy-making
Adam Wyner and Tom van Engers
Submitted to eGOV 2010
Abstract
The paper motivates and proposes a framework for enriched on-line discussion forums for e-government policy-making, where pro and con statements for positions are structured, recorded, represented, and evaluated. The framework builds on current technologies for multi-threaded discussion lists by integrating modes, natural language processing, ontologies, and formal argumentation frameworks. With modes other than the standard reply “comment”, users specify the semantic relationship between a new statement and the previous statement; the result is an argument graph. Natural language processing with a controlled language constrains the domain of discourse, eliminates ambiguity and unclarity, allows a logical representation of statements, and facilitates information extraction. However, the controlled language is highly expressive and natural . Ontologies represent the knowledge of the domain. Argumentation frameworks evaluate the argument graph and generate sets of consistent statements. The output of the system is a rich and articulated representation of a set of policy statements which supports queries, information extraction, and inference
From Policy-making Statements to First-order Logic
Adam Wyner, Tom van Engers, and Kiavash Bahreini
Submitted to eGOVIS 2010
Abstract
Within a framework for enriched on-line discussion forums for e-government policy-making, pro and con statements for positions are input, structurally related, then logically represented and evaluated. The framework builds on current technologies for multi-threaded discussion, natural language processing, ontologies, and formal argumentation frameworks. This paper focuses on the natural language processing of statements in the framework. A small sample policy discussion is presented. We adopt and apply a controlled natural language (Attempto Controlled English) to constrain the domain of discourse, eliminate ambiguity and unclarity, allow a logical representation of statements which supports inference and consistency checking, and facilitate information extraction. Each of the polity statements is automatically translated into rst-order logic. The result is logical representation of the policy discussion which we can query, draw inferences (given ground statements), test for consistency, and extract detailed information.
Towards Web-base Mass Argumentation in Natural Language
Adam Wyner and Tom van Engers
Submitted to EKAW 2010
Abstract
Within the artificial intelligence community, argumentation has been studied for quite some years now. Despite progress, the field has not yet succeeded in creating support tools that members of the public could use to contribute their views to discussions of public policy. One important reason for that is that the input statements of participants in policy-making discussions are put forward in natural language, while translating the statements into the formal models used by argumentation scientists is cumbersome. These formal models can be used to automatically reason with, query, or transmit domain knowledge using web-based technologies. Making this knowledge explicit, formal, and expressed in a language which a machine can process is a labour, time, and knowledge intensive task. To make such translation and it requires expertise that most participants in policy-making debates do not have. In this paper we describe an approach with which we aim at contributing to a solution of this knowledge acquisition bottle-neck. We propose a novel, integrated methodology and framework which adopts and adapts existing technologies. We use semantic wikis which support mass, collaborative, distributive, dynamic knowledge acquisition. In particular, ACEWiki incorporates NLP tools, enabling linguistically competent users to enter their knowledge in natural language, while yielding a logical form that is suitable for automated processing. In the paper we will explain how we can extend the ACEWiki and augment it with argumentation tools which elicit knowledge from users, making implicit information explicit, and generate subsets of consistent knowledge bases from inconsistent knowledge bases. To a set of consistent propositions, we can apply automated reasoners, allowing users to draw inferences and make queries. The methodology and framework take a fragmentary, incremental development approach to knowledge acquisition in complex domains.
By Adam Wyner
Distributed under the Creative Commons
Attribution-Non-Commercial-Share Alike 2.0
The IMPACT Project — first two days
As I mentioned in a previous post, I am working in Amsterdam for the next three months on setting up a research project at the Leibniz Center for Law. The focus here is to develop information extract of textual debates (using GATE) and a tool for inputting debates in a structured manner that can be further processed for reasoning.
The official IMPACT Project information on CORDIS.
As part of my contribution, I have two draft papers, written in the spring and summer of 2009, which will be further developed at Leibniz: From Arguments in Natural Language to Argumentation Frameworks and Multi-modal Multi-threaded Online Forums. While these are early drafts of papers and not for wider circulation, they give a good indication of the line of thinking and of some of the key ideas we will be pursuing. Comments about these works are very welcome.
By Adam Wyner
Distributed under the Creative Commons
Attribution-Non-Commercial-Share Alike 2.0
Forthcoming Article: On Controlled Natural Languages: Properties and Prospects
I am a co-author of the forthcoming article On Controlled Natural Languages: Properties and Prospects. From the abstract:
This collaborative report highlights the properties and prospects of Controlled Natural Languages (CNLs). The report poses a range of questions concerning the goals of the CNL, the design, the linguistic aspects, the relationships and evaluation of CNLs, and the application tools. In posing the questions, the report attempts to structure the field of CNLs and to encourage further systematic discussion by researchers and developers.
The reference and link to the article:
A. Wyner, K. Angelov, G. Barzdins, D. Damljanovic, N. Fuchs, S. Hoefler, K. Jones, K. Kaljurand, T. Kuhn, M. Luts, J. Pool, M. Rosner, R. Schwitter, and J. Sowa. On Controlled Natural Languages: Properties and Prospects, to appear in: N.E. Fuchs (ed.), Workshop on Controlled Natural Languages, CNL 2009, LNCS/LNAI 5972, Springer, 2010.