Dan Corbett

A basic mathematical framework for conceptual graphs

Based on the original idea of Sowa on conceptual graph and a recent formalism by Corbett on ontology, this paper presents a rigorous mathematization of basic concepts encountered in the conceptual structure theory, including canon, ontology, conceptual graph, projection, and canonical formation operations, with the aim of deriving their mathematical properties and applying them to future research and development on knowledge representation. Our proposed formalism enhances the conceptual structure theory and enables it to compare favorably with other alternative methods such as the formal concept analysis theory.

Building corporate knowledge through ontology integration

This paper presents an approach for building corporate knowledge, defined as the total knowledge acquired by an enterprise in its business dealings, through integration of its existing ontologies. We propose to represent corporate knowledge as the final merged ontology, defined under our formalism, in which a canon, or common ontology, is used as the standard under which all other ontologies are aligned. The canon is also enriched with knowledge gained during each ontology merging exercise. Our method ensures that all resulting ontologies are semantically consistent, compact and complete, as well as mathematically sound, so that formal reasoning could be conducted.

An Ontology Formalization of Relation Type Hierarchy in Conceptual Structure Theory

This paper presents an enhancement to ontology formalization, combining previous work in Conceptual Structure Theory and Order-Sorted Logic. In particular, the relation type hierarchy of the former theory is a subset of the predicate hierarchy of the latter. Most existing ontology formalisms place greater importance on concept types, but this paper focuses more on relation types, which are in essence predicates on concept types. New notions are introduced and new properties identified with the aim of completing missing arguments in relation types. The end result is a new ontology, that we call the closure of the original ontology, on which automated inference could be more easily produced (e.g., a query-answering system for legal knowledge).

Meta-relation and ontology closure in conceptual structure theory

This paper presents an enhanced ontology formalization, combining previous work in Conceptual Structure Theory and Order-Sorted Logic. Most existing ontology formalisms place greater importance on concept types, but in this paper we focus on relation types, which are in essence predicates on concept types. We formalize the notion of ‘predicate of predicates’ as meta-relation type and introduce the new hierarchy of meta-relation types as part of the ontology definition. The new notion of closure of a relation or meta-relation type is presented as a means to complete that relation or meta-relation type by transferring extra arguments and properties from other related types. The end result is an expanded ontology, called the closure of the original ontology, on which automated inference could be more easily performed. Our proposal could be viewed as a novel and improved ontology formalization within Conceptual Structure Theory and a contribution to knowledge representation and formal reasoning (e.g., to build a query-answering system for legal knowledge).

Reasoning about time in a BDI architecture

The philosophical roots of the Belief-Desire-Intention model lie in Bratmans formulation of an intention theory of planning, in which he sought to make sense of the notion of future-directed intention. Implementations of BDI mainly follow the original Procedural Reasoning System model. BDI has a sound logical basis, exemplified by the Logic Of Rational Agents. While the LORA formulation has a temporal logic component, however, this does not translate into any ability for the agent to reason about actual time. Being able to reason about actual time would bring significant benefits for BDI agents, such as the ability for agents to communicate deadlines and to plan and schedule activities in a cooperating group. Given a suitable representation of temporal knowledge, an agent could learn about the temporal aspects of its own actions and processes, and this knowledge could be used as input to the planning process. This paper outlines a possible implementation strategy for the representation of, and the capacity to reason about, actual time.

SeseiOnto: Interfacing NLP and Ontology Extraction

For many years, information retrieval tools have been used to try to solve the information overload problem which was accentuated by the coming of age of the World Wide Web. Some tools used Boolean search, others, natural language based processing (NLP). Ontology-based techniques were proposed to improve the quality of the search but none were widely adopted since they did not statistically enhance either the recall or the precision of the search. However, when it comes to information extraction, they may be of significant help. Their integration in professional search engines has been rather slow, partially due to the fact that the ontology building process is time consuming. In this paper, we describe the SeseiOnto software, which uses simple artificial intelligence techniques to improve information extraction and retrieval. To assist the NLP-based information retrieval on a corpus of documents, SeseiOnto employs an automatically generated ontology. Under our experiments, we found that SeseiOnto obtained results comparable to a traditional search engine, while providing a natural language interface to its user

Representing Event Assertions in an Upper Event Ontology

This paper presents an upper event ontology under Conceptual Structure Theory with an emphasis on formalizing the event relation type hierarchy of the ontology to enable representation and classification of event assertions. The proposed ontology essentially consists of an event type hierarchy created from basic event assertions, an event relation type hierarchy built from predicates on event types and other formal relationships between these structures. Representation of event assertions enables inference on events, which can be used in the Semantic Web for automated inference and query answering.

Self Optimization using Conceptual Graphs for NASA Autonomous Systems

NASA is pursing ever increasing autonomous systems to perform new science and exploration. These missions will be out of contact with mission control for extended periods of time in very harsh and unforgiving environments. To successfully perform these missions autonomously, they will have to have autonomic properties to achieve their mission goals. One of theses properties, self-optimization, will be vital to these missions since deep space science and exploration by their nature require operations in unknown environments. Coupling this with the autonomous nature of these missions, they will require a means to optimize themselves as the mission progresses and the actual environment becomes better known. This paper introduces a self-optimization technique that is being developed at SAIC based on conceptual graphs. In this paper an overview of conceptual graphs will be given and how they can be used in autonomous systems to achieve self-optimization

Concept type hierarchy as ontology: an example historical knowledge base

In this paper we explore the issue of using some aspects of the Conceptual Graph Theory formalism to define functions on ontologies We exploit the formal definitions of type hierarchy and projection to define operations on an ontology, and then illustrate these ideas by demonstrating a knowledge base of historical interactions that was implemented on an ontology defined in this way.