Explanation aware computing

This paper takes on the task of providing a formal system of dialogue CE in which the speech acts of requesting and providing an explanation are represented as dialogue moves in the system. CE has opening rules, locution rules, dialogue rules, success rules and closing rules. The system is meant to be simple and basic, to provide a platform for developing more specialized formal dialogue systems of explanation used for specific purposes. The dialogical theory of explanation postulates that an explanation is a dialogue between two parties, one of whom asks a question requesting understanding of something which he or she claims not to understand, while the other offers a response that claims to convey the requested understanding to the party asking the question.

This study describes a discourse approach to explanation aware knowledge representation. It presents a reasoning model that adheres to argumentation as found in written discourse, intended for use in intelligent human-computer collaboration and inter-agent deliberation. The approach integrates the Toulmin model with Rhetorical Structure Theory and Perelman and Olbrechts-Tyteca’s (1958) strategic forms of argumentative processes to define a set of constraints for governing argumentative interactions and formulating explanations in an ontologically normalized manner. Arguments, when satisfied, are instantiated into a dynamic rhetorical network that represents the system’s model of the situation. Two modalities of instantiation are proposed. Inferential instantiation is used when a claim may be inferred from a ground, and synthetic instantiation is used for descriptive argumentation where both ground and claim must be satisfied for the argument to be instantiated. The instantiation pro...

In this paper, we describe precedent-based explanations for case-based classification systems. Previous work has shown that explanation cases that are more marginal than the query case, in the sense of lying between the query case and the decision boundary, are more convincing explanations. We show how to retrieve such explanation cases in a way that requires lower knowledge engineering overheads than previously. We evaluate our approaches empirically, finding that the explanations that our systems retrieve are often more convincing than those found by the previous approach. The paper ends with a thorough discussion of a range of factors that affect precedent-based explanations, many of which warrant further research.

Instructors play a major role in many of the current virtual training systems. Consequently, either many instructors per trainee are needed, which is expensive, or single instructors perform highly demanding tasks, which might lead to suboptimal training. To solve this problem, this paper proposes a cognitive model that not only generates the behaviour of virtual characters, but also produces explanations about it. Agents based on the model reason with BDI concepts, which enables the generation of explanations that involve the underlying reasons for an agent's actions in 'human' terms such as beliefs, intentions, and goals.

The number of news reports published online is now so great that it is impossible for any person to read all of them. Not all of these reports are equally interesting. Automating the identification and evaluation of interest in news is therefore a potentially valuable goal. Traditional ...