this paper we discuss Conceptual Knowledge Discovery in Databases (CKDD) as it is developing in the field of Conceptual Knowledge Processing (cf. [29],[30]). Conceptual Knowledge Processing is based on the mathematical theory of Formal Concept Analysis which has become a successful theory for data analysis during the last 18 years. This approach relies on the pragmatic philosophy of Ch. S. Peirce [15] who claims that we can only analyze and argue within restricted contexts where we always rely on pre-knowledge and common sense. The development of Formal Concept Analysis led to the software system TOSCANA, which is presented as a CKDD tool in this paper. TOSCANA is a flexible navigation tool that allows dynamic browsing through and zooming into the data. It supports the exploration of large databases by visualizing conceptual aspects inherent to the data. We want to clarify that CKDD can be understood as a human-centered approach of Knowledge Discovery in Databases. The actual discussion about human-centered Knowledge Discovery is therefore briefly summarized in Section 1

Abstract not found

Formal Concept Analysis is based on a formalization of context. Since there are situations where the consideration of one context is not sufficient, it is desirable to introduce a formalization of a network of contexts. In this paper, such formalization is given by the notion of multicontext. The aim of the paper is to offer a first study of multicontexts and their conceptual structures; in particular, descriptions of conceptual coherences within the formalized network of contexts are presented. The theoretical considerations are illustrated by examples.

This paper presents the results of experiments carried out with a set of 4,000 medical discharge summaries in which were recognised 1,962 attributes from the Unified Medical Language System (UMLS). In this domain, the objects are medical documents (4,000) and the attributes are UMLS terms extracted from the documents (1,962). When Formal Concept Analysis is used to iteratively analyse and visualize this data, complexity and scalability become critically important. Although the amount of data used in this experiment is small compared with the size of primary memory in modern computers, the results are still important since the probability distributions which determine the efficiencies are likely to remain stable as the size of the data is increased. Our work presents two outcomes, firstly we present a methodology for exploring knowledge in text documents using Formal Concept Analysis by employing conceptual scales created as the result of direct manipulation of a line diagram. The conceptual scales lead to small derived purified contexts that are represented using nested line diagrams. Secondly, we present an algorithm for the fast determination of purified contexts from a compressed representation of the large formal context. Our work draws on existing encoding and compression techniques to show how rudimentary data analysis can lead to substantial efficiency improvements to knowledge visualisation. c fl 1993 Blackwell Publishers, 238 Main Street, Cambridge, MA 02142, USA, and 108 Cowley Road, Oxford, OX4 1JF, UK. SCALABILITY IN FORMAL CONCEPT ANALYSIS 3

Abstract. For over a decade, work on Formal Concept Analysis has been accompanied by the development of the Toscana software. Toscana was implemented to realize the idea of Conceptual Information Systems which allow the analysis of data using concept-oriented methods. Over the years, many ideas from Formal Concept Analysis have been tested in Toscana systems while the real-world problems encountered led to new theoretical research. After ten years of development, the ToscanaJ project was initiated to solve some outstanding problems of the older Toscana versions. The ToscanaJ suite provides programs for creating and using Conceptual Information Systems. The experience with older Toscana implementations has been applied to the design of the programs. A workflow that developed through many Toscana projects has now been integrated into the tools to make them easier to use. Implemented as an Open-Source project and embedded into the larger Tockit project, ToscanaJ is also a starting point for creating a common base for software development for Formal Concept Analysis. In this paper, we present the features of the ToscanaJ suite and how they can be used to implement Conceptual Information Systems. 1

A Conceptual Information System consists of a database together with conceptual hierarchies. The visualization of arbitrary combinations of conceptual hierarchies by nested line diagrams allows an on-line interaction with a database to analyze data conceptually. The chapter describes the conception of Conceptual Information Systems and discusses the use of their visualization techniques for On-Line Analytical Processing (OLAP).

Common clustering techniques have the disadvantage that they do not provide intensional descriptions of the clusters obtained. Conceptual Clustering techniques, on the other hand, provide such descriptions, but are known to be rather slow. In this paper, we discuss a way of combining both techniques. We first cluster the documents by a variant of #--Means, using a thesaurus as background knowledge. This clustering reduces the large number of documents to a relatively small number of clusters, which can then be clustered conceptually in the second step.

While most data processing tools use numerical aspects of the data, conceptual data systems focus on their conceptual structure.

Formal concept analysis (FCA) [10] is a technique for knowledge representation and unsupervised machine learning that has been applied to a wide variety of knowledge representation and exploration tasks in a disparate range of disciplines including civil engineering[7], psychology[8], and political science[9]. Several software tools have been developed to aid knowledge exploration using FCA. The most prevelant of these being TOSCANA[4] but also others[6, 2]. In these tools the process of creating conceptual scales is performed before the analysis stage by experts in FCA after interviews with domain experts. This paper describes a tool to allow the creation of conceptual scales by a domain expert through the manipulation of a folding Hasse diagram. By providing the domain expert with a set of simple graph manipulations, an intuitive understanding of FCA can facilitate the creation of conceptual scales and navigation of the data. 1 Introduction The purpose of the software de...

A Conceptual Information System consists of a database together with conceptual hierarchies. The management system TOSCANA visualizes arbitrary combinations of conceptual hierarchies by nested line diagrams and allows an on-line interaction with a database to analyze data conceptually. The paper describes the conception of Conceptual Information Systems and discusses the use of their visualization techniques for On-Line Analytical Processing (OLAP).