Ebook: Computational Models of Argument

Argumentation has long been a topic of study in Philosophy, and over the last 20 years has become a well established topic in Computer Science, as reflected by this conference on Computational Models of Argument (COMMA). While work on argumentation is broad, the COMMA community is distinguished by its focus on the computational and mathematical aspects of the subject. This focus aims to ensure that its methods are sound – that they, for example, identify arguments that are correct in some sense – but it is also in order to provide an unambiguous specification for implementation, to produce computer programs that reason in that correct way. Much of the work in the COMMA community is prescriptive, concerned with the right way to carry out argument. Another important strand is concerned with building systems that are capable of natural argument, i.e., that will be recognised and found persuasive by humans. Yet another strand that is beginning to emerge is concerned with recognising argument, allowing human arguments to be extracted from text or speech.

The first COMMA conference was held in September 2006 at the University of Liverpool in the United Kingdom, and the conference has been held every two years since. The second edition of the conference was held in May 2008 at the Institut de Recherche en Informatique de Toulouse in France. The third edition was held at Desenzano del Garda in Italy in September 2010, and was organised by the University of Brescia. The fourth edition of the conference was held in September 2012 at the Vienna University of Technology in Austria, and this fifth edition is being held in September 2014 in the beautiful surroundings of Pitlochry, Scotland.

Over the eight years that the conference has been running, it has developed a strong following. The number of submissions has grown to a healthy number (around 50 in 2006, 60 in 2008, 67 in 2010, 76 in 2012 and 67 again this year, 10 of which were submitted to the innovative applications track), and we have researchers who first came to COMMA to present their PhD work who are now submitting papers with their own PhD students. It is particularly gratifying to see the way in which demonstrations – which point the way to practical applications of the techniques we develop – have become an integral part of the conference. This year we have 21 demonstrations, with 19 of these being represented in this proceedings by an extended abstract (the other 2 are associated with another paper in the proceedings).

The selection of papers and demonstrations was made on the basis of the reviews made by members of the Programme Committee and the discussions that followed from those reviews. The quality of every scientific event depends on such work, and we are grateful to the members of the Programme Committee for all the effort that they put into the review process. Thanks to their hard work, we are confident that the papers this year fully deserve to be here. In total we have 24 long papers and 18 short papers (two long and five short in the innovative applications track) each of which will be presented in a plenary talk. These talks will make up the bulk of the technical sessions, complemented by invited talks by Diane Litman, Guillermo Simari and Rineke Verbrugge.

We gratefully acknowledge financial support from the Scottish Informatics and Computer Science Alliance.

Aberdeen/Dundee/Liverpool, July 2014

Simon Parsons (Program Chair)

Nir Oren (Conference Chair)

Chris Reed (Conference Chair)

Federico Cerutti (Demonstrations Chair)

The problem of finding properties that characterizes the relation between argumentation semantics and argumentation inference operators has beginning to surface in the last years. Several works have addresses this concern proposing different “postulates” that reflect the intuitions in this respect. Argumentative reasoning is by nature defeasible and that distinct feature must have a deep influence on the resulting constrains. We believe that the very essence of argumentation should affect the manner in which the required properties are described.

In this paper, we show two important connections between computational models of narrative and computational models of argumentation. First, we show how argumentation techniques can be applied to enrich story understanding, especially where an understanding the story requires understanding of the motives of its characters. This also helps to explain how stories can themselves be seen as as arguments for a particular ordering on values within a value based argumentation framework. We illustrate our discussion using biblical parables, taking as our main example the parable of the Good Samaritan.

The aim of the paper is to show the potential of applying computational models of argument in the financial domain to the analysis of Earnings Conference Calls. We propose the formal description of ECC dialogue games: its locution rules and protocol. We also test its descriptive validity on an annotated corpus.

This paper presents a novel application of argumentation for automated Story Comprehension (SC). It uses argumentation to develop a computational approach for SC as this is understood and studied in psychology. Argumentation provides uniform solutions to various representational and reasoning problems required for SC such as the frame, ramification, and qualification problems, as well as the problem of contrapositive reasoning with default information. The grounded semantics of argumentation provides a suitable basis for the construction and revision of comprehension models, through the synthesis of the explicit information from the narrative in the text with the implicit (in the reader's mind) common sense world knowledge pertaining to the topic(s) of the story given in the text. We report on the empirical evaluation of the approach through a prototype system and its ability to capture both the majority and the variability of understanding of stories by human readers. This application of argumentation can provide an important test-bed for the more general development of computational argumentation.

In this paper, we develop a linguistic and conceptual analysis of argument coumpounds, i.e. groups of closely related arguments and their related contexts expressed by means of discourse relations. An implementation in Dislog is presented with an indicative evaluation of the results.

The constant emergence and change of technologies in the form of digital products and services can cause certain groups of the population to feel excluded, older adults represent one such group. We investigate how to combine applied research on computational models of argument and human-centric computing to impact the way in which older adults interact with broadcast debates. This paper describes a technology application that uses a speech recognition interface to interact with broadcast debates. The application classifies spoken utterances and creates positive or negative “votes” related arguments from a debate. We describe a user study where older adults interact with a debate using the application. Our results indicate that the use of speech recognition, extra information provided to users, and feedback on their interaction, plays an important role in their engagement with the debate.

We present a dialogue game representation of Lakatos's theory of justification and discovery in mathematics and describe our implementation of the game. As well as demonstrating a new domain for dialogue games, this could provide the basis for systems which can be used to aid mathematicians in their work.

The increase in routine clinical data collection coupled with an expectation to exploit this in support of evidence based decision making creates the requirement for a system to support clinicians in this analysis. This paper looks at applying argumentation to this problem, by collating all the relevant statistical approaches and their assumptions into a statistical knowledge base and then representing the model selection process through argumentation. This will form the foundation for the development of a prototype that will enable clinicians to answer their research questions with no statistics, informatics or administrative support.

We introduce a preferential-based setting for reasoning with different types of argumentation-based semantics, including those that are not necessarily conflict-free or admissible. The induced entailments are defined by n-valued labeling and may be computed by answer-set programs.

We introduce a general approach for representing and reasoning with argumentation-based systems. In our framework arguments are represented by Gentzen-style sequents, attacks (conflicts) between arguments are represented by sequent elimination rules, and deductions are made by dynamic proof systems. This framework accommodates different languages and logics in which arguments may be represented, supports a variety of attack relations, and tolerates dynamic changes in the argumentation setting by revising derivations of assertions in light of new information.

We address the problem of explaining Boolean Conjunctive Query (BCQ) failure in the presence of inconsistency within the Ontology-Based Data Access (OBDA) setting, where inconsistency is handled by the intersection of closed repairs semantics (ICR) and the ontology is represented by Datalog+/− rules. Our proposal relies on an interactive and argumentative approach where the processes of explanation takes the form of a dialogue between the User and the Reasoner. We exploit the equivalence between argumentation and ICR-semantics to prove that the Reasoner can always provide an answer for user's questions.

In this paper we extend an argumentation scheme for practical reasoning with values based on Action-based Alternating Transition Systems. While the original scheme considers only arguments arising from the immediately next state, our proposals will enable long term considerations to be taken into account. We consider the various reasons for and against performing an action that arise from these longer term considerations, and develop a new set of argumentation schemes for practical reasoning which allows a clearer separation between facts, values and preferences, and more precise targeting of attacks.

Epistemic probabilities in argumentation frameworks are meant to represent subjective degrees of belief in the acceptance of arguments. As such, they are subject to some rationality conditions, taking into account the attack relation between arguments. This paper provides an advancement with respect to the previous literature on this matter by casting epistemic probabilities in the context of de Finetti's theory of subjective probability and by analyzing and revising the relevant rationality properties in relation with de Finetti's notion of coherence. Further, we consider the extension to Walley's theory of imprecise probabilities and carry out a preliminary analysis about rationality conditions in this more general context.

The process of proof is one of inference to the best explanation, in which alternative scenarios are supported and attacked by arguments. This combination of scenarios and arguments was previously presented as a formal hybrid theory. In this paper, the aim is to further integrate scenarios and arguments by defining a notion of attack between alternative explanations. Thus, scenarios and arguments can be incorporated in the same dialectical framework.

In this paper, we present the Dialogue Game Execution Platform (DGEP), which is able to process and execute any dialogue game specified in a general description language and build arguments and dialogue histories in the language of the AIF ontology. Thus, DGEP allows us to generalise techniques for generation and investigation of dialogues and, through a set of web services, connect a wide variety of multi-agent systems and human-computer interfaces for dialogue.

In this paper we test four different implementations of reasoning tools dedicated to Abstract Argumentation Frameworks. These systems are ASPARTIX, dynPARTIX, Dung-O-Matic, and ConArg2. The tests are executed over three different models of randomly-generated graphs, i.e., the Erdős-Rényi model, the Kleinberg small-world model, and the scale-free Barabasi-Albert model. We compare these four tools with the purpose to test the search of all the possible stable extensions. Then we benchmark dynPARTIX and ConArg2 on the credulous and skeptical acceptance of arguments. Finally, we also evaluate ConArg2 to check the existence of a stable extension.

In this work we address the issue of uncertainty in abstract argumentation. We propose a way to compute the relative relevance of arguments by merging the classical argumentation framework proposed in [5] into a game theoretic coalitional setting, where the worth of a collection of arguments can be seen as the combination of the information concerning the defeat relation and the preferences over arguments of a “user”. Via a property-driven approach, we show that the Shapley value [17] for coalitional games defined over an argumentation framework, can be applied to resume all the information about the worth of sets of arguments into an attribution of relevance for the single arguments. We also prove that, for a large family of (coalitional) argumentation frameworks, the Shapley value can be easily computed.

In an abstract argumentation framework, there are often multiple plausible ways to evaluate (or label) the status of each argument as accepted, rejected, or undecided. But often there exists a critical set of arguments whose status is sufficient to determine uniquely the status of every other argument. Once an agent has decided its position on a critical set of arguments, then essentially the entire frame-work has been evaluated. Likewise, once a group, e.g. a jury, agrees on the status of a critical set of arguments, all of their different views over all other arguments are resolved. Thus, critical sets of arguments are important both for efficient evaluation by individual agents and for collective agreement by groups of such. To exploit this idea in practice, however, a number of computational questions must be considered. In particular, how much computational effort is needed to verify that a set is, indeed, a critical set or a minimal critical set. In this paper we determine exact bounds on the computational complexity of these and related questions. In addition we provide similar analyses of issues: a concept closely related to critical set and derived in terms of (equivalence) classes of arguments related through “common” labelling behaviours.

Argument mining has started to yield early results in automatic analysis of text to produce representations of reason-conclusion structures. This paper addresses for the first time the question of automatically extracting such structures from dialogical settings of argument. More specifically, we introduce theoretical foundations for dialogical argument mining as well as show the initial implementation in a software for dialogue processing, and the application in corpus analysis. We combine analysis of illocutionary structure with structured argumentation frameworks as our scaffolding, and apply a combination of statistical and grammatically based analytical techniques.

In the current paper, we re-examine the concept of strong admissibility, as was originally introduced by Baroni and Giacomin. We examine the formal properties of strong admissibility, both in its extension-based and in its labelling-based form. Moreover, we show that strong admissibility plays a vital role in discussion-based proof procedures for grounded semantics. In particular it allows one to compare the performance of alternative dialectical proof procedures for grounded semantics, and obtain some remarkable differences between the Standard Grounded Game and the Grounded Persuasion Game.

The work of Caminada & Amgoud presents two possible ways of satisfying the rationality postulates: one using restricted rebut, and one using unrestricted rebut. Subsequent work on ASPIC+ has extended the work of Caminada & Amgoud, for instance by allowing preferences over arguments. However, such extensions have utilised restricted rebut only. In the current paper, we make the case for unrestricted rebut, and provide a formalism (called ASPIC−) that implements preferences between the defeasible rules, in the context of unrestricted rebut while still satisfying the rationality postulates of Caminada & Amgoud.