Pml 2: A Modular Explanation Interlingua

2003

Abstract The World Wide Web lacks support for explaining information provenance. When web applications return results, many users do not know what information sources were used, when they were updated, how reliable the source was, what information was looked up versus derived, and if something was derived, how it was derived. In this paper we introduce the Inference Web (IW) that addresses the problems of opaque query answers by providing sharable, combinable, and distributed explanations.

A user of a Semantic Web application may not trust its results because he may not understand how the application produces its results using distributed data and inferential capabilities. Explanation-aware Semantic Web applications provide explanations of their reasoning-explaining why an application has performed a given step or which information it has used to derive a new piece of information. However, providing too much and irrelevant information in explanations may overwhelm the users, especially the non-expert users. In this paper, we discuss an approach to explain reasoning over Linked Data. We introduce a vocabulary to describe justifications and we discuss how publishing justifications as Linked Data enables explaining reasoning over Linked Data. Finally, we discuss how to summarize explanations with relevant information taking into account user specified explanation filtering criterion.

… of the American Association for Artificial …, 2003

Abstract In this position-paper, we have outlined our initial explorations into provisioning for Spoken Natural Language Explanations in Inference Web (IW). We present our motivation for considering speech as an effective modality for presenting explanations, in terms of the interesting usecases that we believe are enabled by the addition of this modality. We then describe the design and implementation of our prototype system. We highlight the salient features of our approach using an illustrative running example.

2012

jcel is a reasoner for the description logic EL + that uses a rule-based completion algorithm. These algorithms are used to get subsumptions in the lightweight description logic EL and its extensions. One of these extensions is EL + , a description logic with restricted expressivity, but used in formal representation of biomedical ontologies. These ontologies can be encoded using the Web Ontology Language (OWL), and through the OWL API, edited using the popular ontology editor Protégé. jcel implements a subset of the OWL 2 EL profile, and can be used as a Java library or as a Protégé plug-in. This system description presents the architecture and main features of jcel, and reports some of the challenges and limitations faced in its development.

2012

Semantic Web applications use interconnected distributed data and inferential capabilities to compute their results. The users of Semantic Web applications might find it difficult to understand how a result is produced or how a new piece of information is derived in the process. Explanation enables users to understand the process of obtaining results. Explanation adds transparency to the process of obtaining results and enables user trust on the process. The concept of providing explanation has been first introduced in expert systems and later studied in different application areas. This paper provides a brief review of existing research on explanation in the Semantic Web.

2020

Interest in the field of Explainable Artificial Intelligence has been growing for decades and has accelerated recently. As Artificial Intelligence models have become more complex, and often more opaque, with the incorporation of complex machine learning techniques, explainability has become more critical. Recently, researchers have been investigating and tackling explainability with a user-centric focus, looking for explanations to consider trustworthiness, comprehensibility, explicit provenance, and context-awareness. In this chapter, we leverage our survey of explanation literature in Artificial Intelligence and closely related fields and use these past efforts to generate a set of explanation types that we feel reflect the expanded needs of explanation for today's artificial intelligence applications. We define each type and provide an example question that would motivate the need for this style of explanation. We believe this set of explanation types will help future system ...

The Semantic Web- …, 2010

Artificial Intelligence, 1980

In this paper we describe the design of BLAH, a system which combines a reasoning capability with an explanation capability. The primary focus of this work is on structuring explanations so that they do not appear complex, and thus are easy to understand. The features of explanation structure that are identified as important include constraints of syntactic form, ways of managing the embedding of explanations, and how the focus of attention is located and shifted. The reasoning component has been designed to take account of certain features needed in generating an acceptable explanation, such as the user's expectations.