Incremental Concept Formation Algorithms Based on Galois (Concept) Lattices (original) (raw)

Incremental concept formation algorithms based on galois lattices

1991

The Galois (or concept) lattice produced from a binary relation has been proved useful for many applications. Building the Galois lattice can be considered as a conceptual clustering method since it results in a concept hierarchy. This article presents incremental algorithms for updating the Galois lattice and corresponding graph, resulting in an incremental concept formation method. Different strategies are considered based on a characterization of the modifications implied by such an update. Results of empirical tests are given in order to compare the performance of the incremental algorithms to three other batch algorithms. Surprisingly, when the total time for incremental generation is used, the simplest and less efficient variant of the incremental algorithms outperforms the batch algorithms in most cases. When only the incremental update time is used, the incremental algorithm outperforms all the batch algorithms. Empirical evidence shows that, on the average, the incremental update is done in time proportional to the number of instances previously treated. Although the worst case is exponential, when there is a fixed upper bound on the number of features related to an instance, which is usually the case in practical applications, the worst case analysis of the algorithm also shows linear growth with respect to the number of instances.

Concept formation by incremental conceptual clustering

1989

Abstract Incremental conceptual clustering is an important area of machine learning. It is concerned with summarizing data in a form of concept hierarchies, which will eventually ease the problem of knowledge acquisition for knowledge-based systems. In this paper we have described INC, a program that generates a hierarchy of concept descriptions incrementally. INC searches a space of classification hierarchies in both top-down and bottom-up fashion.

AN INCREMENTAL CONCEPT-FORMATION APPROACH FOR LEARNING FROM DATABASES

This paper describes a concept formation approach to the discovery of new concepts and implication rules from data. This machine learning approach is based on the Galois lattice theory, and starts from a binary relation between a set of objects and a set of properties (descriptors) to build a concept lattice and a set of rules. Each node (concept) of the lattice represents a subset of objects with their common properties.

Algorithms for Building Concept Sets and Concept Lattices

In our days there is an increasing interest on the application of concept lattices for data mining, especially for generating association rules. The building of concept lattice consists of two, usually distinct phases. In the first phase the set of concepts is generated. The lattice is built in the second phase from the generated set. The paper gives an overview of the available methods and presents a proposed method for contexts of large size where the full context can not be stored in the main memory and some objects may be repeated in the context several times. The proposed algorithm for concept set generation is a fine-tuned version of the incremental concept set building method. At the end of the paper, the test results for comparing the new method with some known methods are given. The proposed method yields in a significantly better cost value than the other methods under the assumed conditions.

Knowledge acquisition via incremental conceptual clustering

Machine Learning, 1987

Conceptual clustering is an important way of summarizing and explaining data. However, the recent formulation of this paradigm has allowed little exploration of conceptual (:lustering as a means of improving performance. Furthermore, previous work in conceptual clustering has not explicitly dealt with constraints imposed by real world environments. This article presents COBWEB, a conceptual clustering system that organizes data so as to maximize inference ability. Additionally, COBWEB is incremental and computationally economical, and thus can be flexibly applied in a variety of domains.

Building Concept (Galois) Lattices from Parts: Generalizing the Incremental Methods

Lecture Notes in Computer Science, 2001

Formal concept analysis and Galois lattices in general are increasingly used as a framework for the resolution of practical problems from software engineering, knowledge engineering and data mining. Recent applications have put the emphasis on the need for both e cient, scalable and exible algorithms to build the lattice. Such features are sought in the development of incremental algorithms. However, the major known incremental algorithm lacks clear theoretical foundations and shows some design aws which strongly a ect its practical performances. Our paper presents a general theoretical framework for the assembly of lattices sharing a same set of attributes based on existing theory on subposition of contexts. The framework underlies the design of a generic procedure for lattice assembly from parts, a new lattice building approach which is more general than the existing incremental and batch ones. As an argument for its theoretical strength, we describe the way our procedure reduces to an improved version of the major known incremental algorithm, which both corrects existing bugs and increases its overall e ciency.

ZooM: a nested Galois lattices-based system for conceptual clustering

Journal of Experimental and Theoretical Artificial Intelligence, 2002

Abstract: This paper deals with the representation of multi-valued databy clustering them in a small number of classes organized in a hierarchyand described at an appropriate level of abstraction. The contributionof this paper is three fold. First we investigate a partial order, namelynesting, relating Galois lattices. A Nested Galois lattice is obtained byreducing (through projections) the original lattice. As a

An incremental concept formation approach to learn and discover from a clinical database

Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94)

The main interest of this research is to discover clinical implications from a large PTCA (Percutaneous Transluminal Coronary Angioplasty) database. A case-based concept formation model D-UNIMEM, modified from Lebowitz's UNIMEM, is proposed for this purpose. In this model, we integrated two kinds of class membership and the index-conjunction class membership. The former is a polythetic clustering approach that serves at the early stage of concept formation. The latter that allows only relevant instances to be placed in the same cluster serves as the later stage of concept formation. D-UNIMEM could extract interesting correlation among features from the learned concept hierarchy.

Towards an Iterative Classification Based on Concept Lattice

Lecture Notes in Computer Science, 2008

In this paper, we propose a generic description of the concept lattice as classifier in an iterative recognition process. We also present the development of a new structural signature adapted to noise context. The experimentation is realized on the noised symbols of GREC database [4]. Our experimentation presents a comparison with the two classical numerical classifiers that are the bayesian classifier and the nearest neighbors classifier and some comparison elements for an iterative process.

Conceptual Clustering: Concept Formation, Drift and Novelty Detection

2010

Abstract. The paper presents a clustering method which can be applied to populated ontologies for discovering interesting groupings of resources therein. The method exploits a simple, yet effective and languageindependent, semi-distance measure for individuals, that is based on their underlying semantics along with a number of dimensions corresponding to a set of concept descriptions (discriminating features committee).