用于辩证分析的多方对话协议

在基于辩论的多agent系统研究中,agent之间的对话博弈一般是双方的,然而现实中的辩论却常常涉及到多方参与者,如何实现多agent系统的多方对话博弈是当前的研究热点之一。用于多方论据博弈的辩证分析模型(DAM-MAG)是一种借鉴中国武术擂台比武思想,将多方对话博弈转化为若干个双方对话博弈的理论模型。DAM-MAG的难点在于多方对话博弈协议的设计和实现。为此,基于该理论模型提出了一种多方对话博弈协议。该协议提供了通过双方对话博弈来解决多方对话博弈问题的方法,为解决多agent系统的多方对话博弈提供了新的途径。     

Enhancing skill in constructing scientific explanations using a structured argumentation scaffold in scientific inquiry

Constructing scientific explanations is necessary for students to engage in scientific inquiry. The purpose of this study is to investigate the influence of using a structured argumentation scaffold to enhance skill in constructing scientific explanations in the process of scientific inquiry. The proposed approach is designed to scaffold the following aspects of argumentation: the argumentation process, the explanation structuring, explanation construction, and explanation evaluation. A quasi-experiment was conducted to examine the effectiveness of the structured argumentation scaffold in developing skill in constructing scientific explanations and engaging in electronic dialogues. A web-based collaborative synchronous inquiry system, ASIS (Argumentative Scientific Inquiry System), was utilized to support students as they worked in groups to carry out inquiry tasks. Two intact sixth grade classes (n = 50) participated in the study. The data show that the ASIS with the structured argumentation scaffold helped students significantly improve their skills in constructing scientific explanations, make more dialogue moves for explanation and query, and use more of all four argument components. In addition, the use of warrants, one of the components of an argument, was found to be a critical variable in predicting students' competence with regard to constructing scientific explanations. The results provide references for further research and system development with regard to facilitating students' construction of scientific argumentation and explanations.     

Visualising knowledge from chat debates in argument diagrams

This study investigates whether combining chat discussion and construction of an argument diagram stimulates students to formulate new ideas in practising argumentation. In this study, 16 secondary school students discussed vivisection and gender equality in pairs using both free and structured chat tools. In structured chat, the students selected and completed partial sentences provided by the computer. After the discussion, they jointly constructed either argument diagrams freely based on the previous discussions with an Internet tool or modified a diagram the computer had constructed automatically during the structured chat. The freely constructed diagrams contained more of the students' prior knowledge than the modified diagrams. However, the different types of diagrams did not differ significantly in breadth, depth, or balance of argumentation. Thus, free construction of argument diagrams seems to activate students to incorporate their prior knowledge into those diagrams.     

基于可信度的辩论模型及争议评价算法

辩论是智能主体间为了消除分歧的一种基于言语的交互行为.由于知识的局限性,争议以及争议内部的陈述通常存在不确定性,因此在对辩论进行建模时需要考虑不确定信息处理问题.提出一种基于可信度的辩论模型（CFA）,该模型将争议表示为由若干前提和一个结论组成的可废止规则,并用对话树描述辩论推演过程.为了表示不确定性推理,引入可信度模型,将争议前提的不确定性和争议之间的攻击强度统一用可信度因子表示.在此基础上,提出计算陈述可信度的争议评价算法,并通过设定可信度阈值确定陈述的可接受性,得出最终辩论结果.最后,用一个实例说明该方法的有效性.该模型可以有效处理不确定信息条件下辩论推理过程,其辩论算法建立在数值计算基础之上,所得出的可接受陈述集在给定可信度阈值条件下是唯一的,可以克服Dung 的抽象辩论框架中扩充语义的不足.     

Argumentation in the Semantic Web

In this article, we define ArgSciFF, a prototype operational argumentation framework to support dialogic argument exchange between Semantic Web services. ArgSciFF is based on the Sciff abductive-logic programming (ALP) framework. (Sciff is an abbreviation for "IFF with constraints for agent societies," referring to the "if and only if' proof procedure developed by Tze Ho Fung and Robert Kowal-ski.) In ArgSciFF, an intelligent agent can interact with a Web service and reason from the interaction result. The reasoning semantics is an argumentation semantics that views the interaction as a dialogue. The dialogue lets two parties exchange arguments and attack, challenge, and justify them on the basis of their knowledge. This format has the potential to overcome a well-known barrier to human users' adoption of IT solutions because it permits interaction that includes justified answers that can be reasoned about and rebutted.     

Is there a Burden of Questioning?

In some recent cases in Anglo-American law juries ruled contrary to an expert's testimony even though that testimony was never challenged, contradicted or questioned in the trial. These cases are shown to raise some theoretical questions about formal dialogue systems in computational dialectical systems for legal argumentation of the kind recently surveyed by Bench-Capon (1997) and Hage (2000) in this journal. In such systems, there is a burden of proof, meaning that if the respondent questions an argument, the proponent is obliged to offer some support for it give it up. But what should happen in a formal system of dialogue if the proponent puts forward an argument and the respondent fails to critically question it, and simply moves on to another issue? Is this some kind of fault that should have implications? Should it be taken to imply that, by default, the respondent has conceded the argument? What, if anything, should be the outcome of such a failure to question in a formal dialogue system of argumentation? These questions are considered by examining some legal cases of expert opinion testimony in relation rules for formal dialectical argumentation systems.     

An inquiry dialogue system

The majority of existing work on agent dialogues considers negotiation, persuasion or deliberation dialogues; we focus on inquiry dialogues, which allow agents to collaborate in order to find new knowledge. We present a general framework for representing dialogues and give the details necessary to generate two subtypes of inquiry dialogue that we define: argument inquiry dialogues allow two agents to share knowledge to jointly construct arguments; warrant inquiry dialogues allow two agents to share knowledge to jointly construct dialectical trees (essentially a tree with an argument at each node in which a child node is a counter argument to its parent). Existing inquiry dialogue systems only model dialogues, meaning they provide a protocol which dictates what the possible legal next moves are but not which of these moves to make. Our system not only includes a dialogue-game style protocol for each subtype of inquiry dialogue that we present, but also a strategy that selects exactly one of the legal moves to make. We propose a benchmark against which we compare our dialogues, being the arguments that can be constructed from the union of the agents’ beliefs, and use this to define soundness and completeness properties that we show hold for all inquiry dialogues generated by our system.     

Partial information basis for agent-based collaborative dialogue

We propose a partial information state-based framework for collaborative dialogue and argument between agents. We employ a three-valued based nonmonotonic logic, NML3, for representing and reasoning about Partial Information States (PIS). NML3 formalizes some aspects of revisable reasoning and it is sound and complete. Within the framework of NML3, we present a formalization of some basic dialogue moves and the rules of protocols of some types of dialogue. The rules of a protocol are nonmonotonic in the sense that the set of propositions to which an agent is committed and the validity of moves vary from one move to another. The use of PIS allows an agent to expand consistently its viewpoint with some of the propositions to which another agent, involved in a dialogue, is overtly committed. A proof method for the logic NML3 has been successfully implemented as an automatic theorem prover. We show, via some examples, that the tableau method employed to implement the theorem prover allows an agent, absolute access to every stage of a proof process. This access is useful for constructive argumentation and for finding cooperative and/or informative answers.     

Combining interaction and context design to support collaborative argumentation using a tool for synchronous CMC

Abstract Empirical studies and theory suggest that educational dialogue can be used to support learners in the development of reasoning, critical thinking and argumentation. This paper presents an educational design for synchronous online peer discussion that guides student dialogue in ways that lead to improved argumentation and collaborative knowledge development. This design includes a mediating interface – or tool, linked to a broader set of online educational activities – a designed local context, where the latter aims to provide conditions that support argumentation. The approach is based on collaborative working and dialogue game approaches to discussion. Preliminary findings with UK Open University students showed the argumentation process was more coherent, varied, deeper and extended when using our interaction design compared with the use of a simple unstructured interface.     

A Dialogue Game Protocol for Agent Purchase Negotiations

We propose a dialogue game protocol for purchase negotiation dialogues which identifies appropriate speech acts, defines constraints on their utterances, and specifies the different sub-tasks agents need to perform in order to engage in dialogues according to this protocol. Our formalism combines a dialogue game similar to those in the philosophy of argumentation with a model of rational consumer purchase decision behaviour adopted from marketing theory. In addition to the dialogue game protocol, we present a portfolio of decision mechanisms for the participating agents engaged in the dialogue and use these to provide our formalism with an operational semantics. We show that these decision mechanisms are sufficient to generate automated purchase decision dialogues between autonomous software agents interacting according to our proposed dialogue game protocol.     

Subjective logic and arguing with evidence

This paper introduces a subjective logic based argumentation framework primarily targeted at evidential reasoning. The framework explicitly caters for argument schemes, accrual of arguments, and burden of proof; these concepts appear in many types of argument, and are particularly useful in dialogues revolving around evidential reasoning. The concept of a sensor is also useful in this domain, representing a source of evidence, and is incorporated in our framework. We show how the framework copes with a number of problems that existing frameworks have difficulty dealing with, and how it can be situated within a simple dialogue game. Finally, we examine reasoning machinery that enables an agent to decide what argument to advance with the goal of maximising its utility at the end of a dialogue.     

PARMENIDES: Facilitating Deliberation in Democracies

Governments and other groups interested in the views of citizens require the means to present justifications of proposed actions, and the means to solicit public opinion concerning these justifications. Although Internet technologies provide the means for such dialogues, system designers usually face a choice between allowing unstructured dialogues, through, for example, bulletin boards, or requiring citizens to acquire a knowledge of some argumentation schema or theory, as in, for example, ZENO. Both of these options present usability problems. In this paper, we describe an implemented system called PARMENIDES which allows structured argument over a proposed course of action, without requiring knowledge of the underlying argumentation theory.     

LASAD: Flexible representations for computer-based collaborative argumentation

Teaching argumentation is challenging, and the factors of how to effectively support the acquisition of argumentation skills through technology are not fully explored yet. One of the key reasons for that is the lack of comparability between studies. In this article, we describe LASAD, a collaborative argumentation framework that can be flexibly parameterized. We illustrate the flexibility of the framework with respect to visualization, structural definitions and kind of cooperation. Using this framework, this paper presents an evaluation of the impact of using an argumentation system with different argument representations and with collaborative vs. individual use on the outcomes of scientific argumentation. We investigate which combinations of these factors produces the best results concerning argument production and learning outcomes. The results of this controlled lab study with 36 participants showed that the use of simple representational formats is superior compared to highly structured ones. Even though the latter encouraged the provision of additional non-given material, the former is less error-prone. A hypothesized structural guidance provided by more complex formats could not be confirmed. With respect to collaboration, the results highlight that arguing in groups lead to more cluttered argumentation maps, including a higher amount of duplicate elements. An expected peer-reviewing between group members did not occur. Yet, groups also tended to include more points-of-view in their arguments, leading to more elaborated argument maps.     

An Empirical Evaluation of Structured Argumentation Using the Toulmin Argument Formalism

Some structured argumentation tools employ the Toulmin argument formalism, but no research has been performed testing this formalism's effect on argument evaluation or communication. An experiment was conducted to address this need by assessing: 1) if the process of generating Toulmin structures impacted participant (re)assessment of the soundness of an argument presented in an article and 2) if other participants thought that the structured representations adequately reflected the written argument. Results were mixed. First, generating Toulmin structures did impact the assessment of argument soundness. This was noteworthy given that participants were professionals representing the population of interest and that a weak manipulation and small sample size were used in the experiment. However, the effect was limited to the article where the argument was poorly aligned with the Toulmin formalism, and second, participants reviewing these structures found them to be less sound than the argument presented in the article itself. More generally, participants did not find it easy to generate Toulmin structures. Greater perceived difficulty in structure generation (and not generation time) was significantly correlated with the amount of change in the participants' soundness ratings, suggesting the mediating role of cognitive effort on reassessment. Generated structures varied greatly. Structures that had more total elements were easier to understand and were given better soundness ratings. These findings suggest that one needs to be cautious of the claimed value of the structured argumentation tools employing the Toulmin formalism without additional empirical research, demonstrating whether and how they can be effective cognitive aids     

Baseballs and arguments from fairness

This paper applies two argumentation schemes, argument from fairness and argument from lack of knowledge (along with other schemes of lesser prominence) to model the reasoning given by Judge McCarthy supporting his decision to divide the proceeds of a homerun baseball in the case of Popov v. Hayashi. Several versions of both schemes are explained and discussed, and then applied to the argumentation given by Judge McCarthy as the basis of the reasoning used to arrive at his decision. The scheme for argument from fairness is shown to be based on a special principle in Perelman’s theory of justice.     

A structured dialogue tool for argumentative learning

This paper presents a structured environment for Computer‐Supported Collaborative Argumentation, which we call the Argumentative Learning Experience (ALEX). The system aims to improve understanding of argumentation and to widen and deepen the space of debate among 16–18‐year‐old students. To use ALEX users make arguments by selecting and completing partial sentences. An automatically created visual representation of the argument is displayed and personalised advice on the argumentation is provided to each user.     

Assisting students with argumentation plans when solving problems in CSCL

In CSCL systems, students who are solving problems in group have to negotiate with each other by exchanging proposals and arguments in order to resolve the conflicts and generate a shared solution. In this context, argument construction assistance is necessary to facilitate reaching to a consensus. This assistance is usually provided with isolated arguments by demand, but this does not offer students a real and integral view of the conflicts. In this work, we study the utilisation of argumentation plans to assist a student during the argumentation. The actions of an argumentation plan represent the arguments that a student might use during the argumentation process. Moreover, these plans can be integrated with the tasks needed to reach a shared solution. These plans give the student an integral and intuitive view of the problem resolution and the conflict that must be resolved. We evaluated our proposal with students of an Artificial Intelligence course. This evaluation was carried out by comparing three different assistance scenarios in which students had to solve exercises: no assistance, assistance with isolated arguments, and assistance with argumentation plans. The results obtained show that reaching consensus was easier for the students when the assistance was provided using argumentations plans.     

Representing and applying knowledge for argumentation in a social context

The concept of argumentation in AI is based almost exclusively on the use of formal, abstract representations. Despite their appealing computational properties, these abstractions become increasingly divorced from their real world counterparts, and, crucially, lose the ability to express the rich gamut of natural argument forms required for creating effective text. In this paper, the demands that socially situated argumentation places on knowledge representation are explored, and the various problems with existing formalisations are discussed. Insights from argumentation theory and social psychology are then adduced as key contributions to a notion of social context which is both computationally tractable and suitably expressive for handling the complexities of argumentation found in natural language.     

Mediating debate through on-line large-scale argumentation: Evidence from the field

Web 2.0 technologies, such as forums and wikis, are enabling an explosion of global knowledge sharing through distributed large-scale conversations, but they seem to be less successful at supporting collaborative deliberation around complex and controversial questions. In order to cope with this limitation, many scholars have proposed to adopt on-line argumentation platforms to improve information visualization, organization and reuse. However, such research has mostly focused on the design of adequate argument-based knowledge formalisms. Less attention has been paid to the empirical analysis of actual interactions mediated by argumentation technology with reasonably large user communities. In this paper, we present an in-depth analysis of the data obtained in the empirical test of an argumentation platform where a 160-member community created, in 3 weeks, what is to our knowledge the largest single online argument map ever built (around 5000 posts). Our results show that (i) users were able to quickly and comprehensively explore and map the debate on the selected discussion topic; (ii) substantial moderation was needed to ensure that the argument map was well-organized and users were confident with the argumentation formalism; (iii) considerable out-of-the map communication occurred, possibly as a way to allow for conversational flows inhibited by the argumentation formalism, (iv) formal rating of contributions favored exploration of the map, understanding the debate structure, and improving the quality of content.     

Dialogue games that agents play within a society

Human societies have long used the capability of argumentation and dialogue to overcome and resolve conflicts that may arise within their communities. Today, there is an increasing level of interest in the application of such dialogue games within artificial agent societies. In particular, within the field of multi-agent systems, this theory of argumentation and dialogue games has become instrumental in designing rich interaction protocols and in providing agents with a means to manage and resolve conflicts. However, to date, much of the existing literature focuses on formulating theoretically sound and complete models for multi-agent systems. Nonetheless, in so doing, it has tended to overlook the computational implications of applying such models in agent societies, especially ones with complex social structures. Furthermore, the systemic impact of using argumentation in multi-agent societies and its interplay with other forms of social influences (such as those that emanate from the roles and relationships of a society) within such contexts has also received comparatively little attention. To this end, this paper presents a significant step towards bridging these gaps for one of the most important dialogue game types; namely argumentation-based negotiation (ABN). The contributions are three fold. First, we present a both theoretically grounded and computationally tractable ABN framework that allows agents to argue, negotiate, and resolve conflicts relating to their social influences within a multi-agent society. In particular, the model encapsulates four fundamental elements: (i) a scheme that captures the stereotypical pattern of reasoning about rights and obligations in an agent society, (ii) a mechanism to use this scheme to systematically identify social arguments to use in such contexts, (iii) a language and a protocol to govern the agent interactions, and (iv) a set of decision functions to enable agents to participate in such dialogues. Second, we use this framework to devise a series of concrete algorithms that give agents a set of ABN strategies to argue and resolve conflicts in a multi-agent task allocation scenario. In so doing, we exemplify the versatility of our framework and its ability to facilitate complex argumentation dialogues within artificial agent societies. Finally, we carry out a series of experiments to identify how and when argumentation can be useful for agent societies. In particular, our results show: a clear inverse correlation between the benefit of arguing and the resources available within the context; that when agents operate with imperfect knowledge, an arguing approach allows them to perform more effectively than a non-arguing one; that arguing earlier in an ABN interaction presents a more efficient method than arguing later in the interaction; and that allowing agents to negotiate their social influences presents both an effective and an efficient method that enhances their performance within a society.