Abstract not found

The aim of this paper is to indicate how TOSCANA may be extended to allow graphical representations not only of concept lattices but also of concept graphs in the sense of Contextual Logic. The contextual-logic extension of TOSCANA requires the logical scaling of conceptual and relational scales for which we propose the Peircean Algebraic Logic as reconstructed by R. W. Burch. As graphical representations we recommend, besides labelled line diagrams of concept lattices and Sowa's diagrams of conceptual graphs, particular information maps for utilizing background knowledge as much as possible. Our considerations are illustrated by a small information system about the domestic ights in Austria.

We describe a knowledge server that permits Web users to retrieve and add knowledge in a shared knowledge base. The following features distinguish WebKB-2 from other ontology servers or KBMSs: (i) the ontology is large (at present, 69,000 categories and 87,800 links mostly coming from WordNet) and extendible at any time by any user, (ii) asynchronous cooperation between users is supported and encouraged (users are encouraged to re-use, complement or correct the knowledge of other users but do not have to agree with each other and may add new names to categories) while the knowledge base is kept unique to maximize knowledge interconnection, retrieval and inconsistency detection, (iii) the proposed knowledge representation languages are designed to be both expressive and readable to permit and encourage the users to enter all the knowledge they want (though that still requires motivation).

This paper presents a CG query engine which is based on the idea of power context families stored in a relational database. It describes the methodology used and aims to demonstrate how this methodology can replace the need of a CG subsumption engine with a simple query engine. It shows how ontologies are used as conceptual scales in order to achieve this goal. It further integrates the WebKB-2 ontology server, addresses the techniques adapted and presents an implementation of the engine in form of a C API.

This paper describes an approach to visualizing Z specifications based on Formal Concept Analysis (FCA). The approach takes a source specification written in L T E X and produces a formal context representing the static structure of the specification. The corresponding formal concept lattice can be used to investigate and explore various properties of the specification. The line diagram does not replace, but is intended to be used in conjunction with, the original Z specification.

This report explores the space of this problem and develops some steps towards its solutions. Our experience with knowledge representation languages and dynamic hyperlinking of HTML documents using conceptual graphs seems complementary to the multiple source knowledge fusion task [40]. The reasons are that conceptual graphs are based on lexical, hierarchically structured ontologies of semantically related terms and these structures lend themselves to traditional information retrival engines, at least so far as term extraction from text is concerned. Furthermore, conceptual graphs provide for typical and necessary conditions, logical inference and term signatures that aid with disambiguiation of semantic intent. Conceptual graphs also permit knowledge fusion by way of the maximal join operation. Conceptual graphs are not without their research challenges however. Who or what can automatically generate the CGs and how the approach scales to real-world multiple source fusion are open questions. To address the first of these issues. There are two schools of thought on who or what generates the CGs needed for knowledge fusion from multiple sources. The first of these is that index expressions generated by a meta-level information retrival engine called the Hyperindex Browser (the HiB) [5, 6, 7, 4] can be used to automatically generate knowledge annotations as input to an inference engine (such as the one used by our research group's WebKB software [38, 39, 40]). The idea here is that we can efficiently extract and fuse knowledge structures from example documents to construct a knowledge exemplar and that these exemplars are subsequently used to compute the closure of the knowledge base and generate pattern recognition classifiers (see Figure 1.1). Domain background knowled...

}) )$ Figure 25: A language for knowledge indexing or connecting any Webaccessible document element. The above indexation notations allow the statements and the indexed DEs to be in different documents. Thus, any user may index any element of a document on the Web. Figure 22 presents a general interface for knowledge-based queries and shows how a document containing knowledge is loaded in the WebKB processor before being queried. 24 KNOWLEDGE PROCESSING THE HARD WAY Peter Eklund WebKB also allows the document owner to index an image by a knowledge statement directly stored in the "alt" field of the HTML "img" tag. We use this special case of indexation to present a simple illustration of WebKB's features. This example, shown in Figure 26, is a good synthesis but is in no way representative of the general use of WebKB --- it is not representative because it mixes the indexed source data (in this case, a collection of images), their indexation, and a customized interface to query th...

Formal Concept Analysis (FCA) is a data analysis technique based on ordered lattice theory. It provides graph-based visualisations of tabular data and has successfully been applied to a number of elds including Text Data Mining, Psychology, Social Science and Software Engineering. This research proposal sets out a framework for the application of FCA to Formal Methods. Formal Methods can be broadly dened as tools and notations that support the unambiguous specication of computer systems and software. While there can be signicant advantages obtained by integrating Formal Methods into the production of software artifacts there is an associated cost. The complexity of Formal Methods means they can be dicult to use and have a non-trivial learning curve. The aim of the research described in this proposal is to increase the accessibility of Formal Methods by providing alternative visual representations of specications through the application of Formal Concept Analysis. ii Contents A...