Industrial Information and Design Issues pp 267-272 | Cite as
Case-Based Reasoning Management of a Structural Database
Abstract
DBMS systems and databases in chemistry, whether for retrieval information or retrieval strategy, are usually limited by their procedural reasoning tools. The limitations of usual systems are often caused by a lack of flexibility due to rigid structuration of the search space, the choice of a given logic and, very often, the fuzzy nature of some of the basic state of space blocks. We are currently investigating the power of analogical search and retrieval tools with an eye to developing a flexible system through the “case history” strategy and the use of “case memory” storing potential similar features chosen by users. The context adaptability parameters, ruling the DBMS reorganization should belong to numerous spaces such as “topology, metrics and physical properties”.
Key words
Artificial intelligence machine learning analogy reasoning case based reasoning substructural search system fuzzy search similarity metrics DARC systemResume
Les systèmes actuels ne supportent usuellement qu’un seul mode de raisonnement: “le raisonnement procédural”. Des progrès importants impliquent différents modes de raisonnement en IA (déductif inductif abductif). Cependant, à cause de la prématurité “technique” d’une telle intégration dans un même système, nous proposons une approche basée sur le raisonnement par analogie “le raisonnement à partir de cas “autour d’un système de recherche structurale floue. L’apprentissage à long terme à partir d’une mémoire des sessions d’interrogation permet une adaptation dynamique des métaparamètres stratégiques de recherche et organisationnels de la base de données à un contexte d’utilisation. Dans le prototype développé sous UNIX nous acquerrons une grande flexibilité en combinant les outils souples du flou générique du système DARC à un programme d’apprentissage par induction à partir de “cas similaires comparés à différents niveaux” (topologie, métrique, propriété).
Preview
Unable to display preview. Download preview PDF.
References
- [1]Forsyth R., Rada R., Machine Learning: Applications in Expert Systems and Information Retrieval, Ellis Horwood Series in AI (1986)Google Scholar
- [2]Chiamella Y., Defude B., A Prototype of an Intelligent System for Information Retrieval: IOTA, Information Processing & Managment System, vol 23 n°4, pp, 285–303 (1987)CrossRefGoogle Scholar
- [3]Slade S., Case-based Reasoning: a Research Paradigm, AI magazine vol 12, n°l, pp 42–53 (1991)Google Scholar
- [4]Chen Z., Enhancing Database Management to Knowledge Base Management: the Role of Information Retrieval Technology, Information Processing & Management system, vol 30, n° 3, pp 419–435 (1994)CrossRefGoogle Scholar
- [5]Rouvray D.H., The Necessity of Analogies in the Development of Science, J. Chem. Inf. Sci., vol 34, pp 446–452 (1994)CrossRefGoogle Scholar
- [6]Golging A.R., Rosenbloom P.S., Improving Rule-Based Systems through Case-Based Reasoning. Knowledge Acquisition and Learning (1993)Google Scholar
- [7]Work done by SYCOGMA, SA ParisGoogle Scholar