PML 2: A Modular Explanation Interlingua

AAAI Conferences

In the past five years, we have designed and evolved an interlingua for sharing explanations generated by various automated systems such as hybrid web-based question answering systems, text analytics, theorem proving, task processing, web services execution, rule engines, and machine learning components.


Explaining Task Processing in Cognitive Assistants That Learn

AAAI Conferences

As personal assistant software matures and assumes more autonomous control of its users' activities, it becomes more critical that this software can explain its task processing. It must be able to tell the user why it is doing what it is doing, and instill trust in the user that its task knowledge reflects standard practice and is being appropriately applied. We will describe the ICEE (Integrated Cognitive Explanation Environment) explanation system and its approach to explaining task reasoning. Key features include (1) an architecture designed for reuse among many different task execution systems; (2) a set of introspective predicates and a software wrapper that extract explanationrelevant information from a task execution system; (3) a version of the Inference Web explainer for generating formal justifications of task processing and converting them to userfriendly explanations; and (4) a unified framework for explanation in which the task explanation system is integrated with previous work on explaining deductive reasoning. Our work is focused on explaining belief-desire-intention (BDI) agent execution frameworks with the ability to learn. We demonstrate ICEE's application within CALO, a state-of-the-art personal software assistant, to explain the task reasoning of one such execution system.


Inference Web: Portable and Sharable Explanations for Question Answering

AAAI Conferences

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. The explanations include information concerning where answers came from and how they were deduced (or retrieved). The IW solution includes: an extensible registry containing details on information sources and reasoners, a portable proof specification, and an explanation browser.


Explaining Task Processing in Cognitive Assistants That Learn

AAAI Conferences

As personal assistant software matures and assumes more autonomous control of user activities, it becomes more critical that this software can explain its task processing. It must be able to tell the user why it is doing what it is doing, and instill trust in the user that its task knowledge reflects standard practice and is being appropriately applied. We will describe the ICEE (Integrated Cognitive Explanation Environment) explanation system and its application to explaining task reasoning, Key features include (1) an architecture designed for reuse among different task execution systems; (2) a set of introspective predicates and a software wrapper that extract explanationrelevant information from a task execution system; (3) a version of the Inference Web explainer for generating formal justifications of task processing and converting them to userfriendly explanations; and (4) a unified framework for explaining results from task execution, learning, and deductive reasoning. Our work is focused on explaining belief-desire-intention (BDI) agent execution frameworks with the ability to learn. We demonstrate ICEE's application within CALO, a state-of-the-art personal software assistant, to explain the task reasoning of one such execution system and describe our associated trust study.


Answering Science Questions: Deduction with Answer Extraction and Procedural Attachment

AAAI Conferences

An approach to question answering through automated deduction is advocated. Answers to questions are extracted from proofs of associated conjectures over an axiomatic theory of the subject domain. External knowledge resources, including data and software, are consulted through a mechanism known as procedural attachment. A researcher ignorant of the subject domain theory or its logical language can formulate questions via a query elicitation facility. A similar device allows an expert to extend the theory. An English explanation for each answer, and a justification for its correctness, is constructed automatically from the proof by which it was extracted. A deductive approach has been applied in planetary astronomy, geography, intelligence analysis, and, most recently, molecular biology and medical research applications. It is argued that the constructs in the Semantic Web languages, including OWL with SWRL, are insufficiently expressive for this kind of application.