Sebastián Gottifredi

A survey of different approaches to support in argumentation systems

Fil: Cohen, Andrea. Universidad Nacional del Sur. Departamento de Ciencias e Ingenieria de la Computacion; Argentina. Consejo Nacional de Investigaciones Cientificas y Tecnicas; Argentina

Relational databases as a massive information source for defeasible argumentation

Argumentation provides a sophisticated yet powerful mechanism for the formalization of commonsense reasoning in knowledge-based systems, with application in many areas of Artificial Intelligence. Nowadays, most argumentation systems build their arguments on the basis of a single, fixed knowledge base, often under the form of a logic program as in Defeasible Logic Programming or in Assumption-Based Argumentation. Currently, adding new information to such programs requires a manual encoding, which is not feasible for many real-world environments which involve large amounts of data, usually conceptualized as relational databases. This paper presents a novel approach to compute arguments from premises obtained from relational databases, identifying several relevant aspects. In our setting, different databases can be updated by external, independent applications, leading to changes in the spectrum of available arguments. We present algorithms for integrating a database management system with an argument-based inference engine. Empirical results and running-time analysis associated with our approach show that it provides a powerful alternative for efficiently achieving massive argumentation, taking advantage of modern DBMS technologies. We contend that our proposal is significant for developing new architectures for knowledge-based applications, such as Decision Support Systems and Recommender Systems, using argumentation as the underlying inference model.

An approach to abstract argumentation with recursive attack and support

This work introduces the Attack-Support Argumentation Framework (ASAF), an approach to abstract argumentation that allows for the representation and combination of attack and support relations. This framework extends the Argumen-tation Framework with Recursive Attacks (AFRA) in two ways. Firstly, it adds a support relation enabling to express support for arguments; this support can also be given to attacks, and to the support relation itself. Secondly, it extends AFRAs attack relation by allowing attacks to the aforementioned support relation. Moreover, since the support relation of the ASAF has a necessity interpretation, the ASAF also extends the Argumentation Framework with Necessities (AFN). Thus, the ASAF provides a unified framework for representing attack and support for arguments, as well as attack and support for the attack and support relations at any level.

Improving argumentation-based recommender systems through context-adaptable selection criteria

An argumentative Reasoning Server is presented.The server can use different comparison criteria changing it dynamically.Properties of the conditional expressions used to select the criterion are studied. Recommender systems based on argumentation represent an important proposal where the recommendation is supported by qualitative information. In these systems, the role of the comparison criterion used to decide between competing arguments is paramount and the possibility of using the most appropriate for a given domain becomes a central issue; therefore, an argumentative recommender system that offers an interchangeable argument comparison criterion provides a significant ability that can be exploited by the user. However, in most of current recommender systems, the argument comparison criterion is either fixed, or codified within the arguments. In this work we propose a formalization of context-adaptable selection criteria that enhances the argumentative reasoning mechanism. Thus, we do not propose of a new type of recommender system; instead we present a mechanism that expand the capabilities of existing argumentation-based recommender systems. More precisely, our proposal is to provide a way of specifying how to select and use the most appropriate argument comparison criterion effecting the selection on the users preferences, giving the possibility of programming, by the use of conditional expressions, which argument preference criterion has to be used in each particular situation.

Query-based argumentation in agent programming

In this work we will present an integration of a query-answering argumentation approach with an abstract agent programming language. Agents will argumentatively reason via special context-based queries, using their mental information. A set of mind affecting capabilities will allow the agent to dynamically modify its mental components and the argument comparison criterion. We will show the models for deliberation and execution of this integrated framework.

Consistent Query Answering Using Relational Databases through Argumentation

This paper introduces a framework that integrates a reasoner based on defeasible argumentation with a large information repository backed by one or several relational databases. In our scenario, we assume that the databases involved are updated by external independent applications, possibly introducing inconsistencies in a particular database, or leading to inconsistency among the subset of databases that refer to the same data. Argumentation reasoning will contribute with the possibility of obtaining consistent answers from the information repository with the properties described. We present the Database Integration for Defeasible Logic Programming (DBI-DeLP) framework, which enables commonsense reasoning based on Defeasible Logic Programming (DeLP) by extending the system capabilities to handle large amounts of data and providing consistent answers for queries posed to it.

A heuristics-based pruning technique for argumentation trees

Argumentation in AI provides an inconsistency-tolerant formalism capable of establishing those pieces of knowledge that can be warranted despite having information in contradiction. Computation of warrant tends to be expensive; in order to alleviate this issue, we propose a heuristics-based pruning technique over dialectical trees. Empirical testing shows that in most cases our approach answers queries much faster than the usual techniques, which prune with no guide.

Sharing beliefs among agents with different degrees of credibility

This paper introduces an approach for sharing beliefs in collaborative multi-agent application domains where some agents can be more credible than others. In this context, we propose a formalization where every agent has its own partial order among its peers representing the credibility the agent assigns to its informants; each agent will also have a belief base where each sentence is attached with an agent identifier which represents the credibility of that sentence. We define four different forwarding criteria for computing the credibility information for a belief to be forwarded, and for determining how the receiver should handle the incoming information; the proposal considers both the sender’s and the receiver’s points of view with respect to the credibility of the source of the information.

Using argument strength for building dialectical bonsai

Argumentation in AI provides an inconsistency-tolerant formalism capable of establishing those pieces of knowledge that can be accepted despite having information in contradiction. Computation of accepted arguments tends to be expensive; in order to alleviate this issue, we propose a heuristics-based pruning technique over argumentation trees. Empirical testing shows that in most cases our approach answers queries much faster than the usual techniques, which prune with no guide. The heuristics is based on a measure of strength assigned to arguments. We show how to compute these strength values by providing the corresponding algorithms, which use dynamic programming techniques to reutilise previously computed trees. In addition to this, we introduce a set of postulates characterising the desired behaviour of any strength formula. We check the given measure of strength against these postulates to show that its behaviour is rational. Although the approach presented here is based on an abstract argumentation framework, the techniques are tightly connected to the dialectical process rather than to the framework itself. Thus, results can be extrapolated to other dialectical-tree-based argumentation formalisms with no additional difficulty.

On influence and contractions in defeasible logic programming

In this paper, we investigate the problem of contraction in Defeasible Logic Programming (DeLP), a logic-based approach for defeasible argumentation.We develop different notions of contraction based on both, the different forms of entailment implicitly existent in argumentation-based formalisms and the influence literals exhibit in the reasoning process. We give translations of widely accepted rationality postulates for belief contraction to our framework. Moreover we discuss on the applicability of contraction for defeasible argumentation and the role of influence in this matter.