Data-driven constructive induction in AQ17-PRE: A method and experiments

Bloedorn, Eric; Michalski, Ryszard S.

doi:10.1109/tai.1991.167073

Cited by 17 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If an adequate model does not lie within that space, then it is not possible for the machine learning system to find one. While constructive induction [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] may construct new complex terms from given primitive terms, it is unable to construct new terms that can only be described using terms that lie outside the given vocabulary. By contrast, human experts are accomplished at extending and refining metamodels as circumstances demand and they have ready access to multiple sources of insight that may help guide such a process.…”

Section: Relative Capabilities Of Machine Learning and Knowledge Acqumentioning

confidence: 99%

Integrating machine learning with knowledge acquisition through direct interaction with domain experts

Webb

1996

Knowledge-Based Systems

View full text Add to dashboard Cite

Knowledge elicitation from experts and empirical machine learning are two distinct approaches to knowledge acquisition with differing and mutually complementary capabilities. Learning apprentices have provided environments in which a knowledge engineer may collaborate with a machine learning system allowing for a synergy between the complementary approaches. The Knowledge Factory is a knowledge acquisition environment that allows a domain expert to collaborate directly with a machine learning system without the need for assistance from a knowledge engineer. This requires a different form of environment to the learning apprentice. The paper describes techniques for supporting such interactions and their implementation in a knowledge acquisition environment called The Knowledge Factory.

show abstract

Section: Relative Capabilities Of Machine Learning and Knowledge Acqumentioning

confidence: 99%

Integrating machine learning with knowledge acquisition through direct interaction with domain experts

Webb

1996

Knowledge-Based Systems

View full text Add to dashboard Cite

show abstract

“…A di erent formalism, decision lists, i n w h i c h the rule ordering is signi cant, will be examined later. Such rules can beconstructed by generalising from positive and negative examples of a concept using the Candidate Elimination algorithm 71] (AQ 66,11,12] The AQ algorithms start with a random single instance (seed) and form the most general hypothesis which includes no negative examples by successively specialising an initially empty complex, using tests that incorporate the seed. Geometrically, they obtain the biggest hyperrectangle enclosing no negative points.…”

Section: Rule-based Systemsmentioning

confidence: 99%

Convex Hulls in Concept Induction

Newlands

Webb

1999

Methodologies for Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Classi cation learning is dominated by systems which induce large numbers of small axis-orthogonal decision surfaces. This strongly biases such systems towards particular hypothesis types but there is reason believe t h a t many domains have underlying concepts which do not involve axis orthogonal surfaces. Further, the multiplicity of small decision regions mitigates against any holistic appreciation of the theories produced by these systems, notwithstanding the fact that many of the small regions are individually comprehensible. This thesis investigates modeling concepts as large geometric structures in n-dimensional space. Convex hulls are a superset of the set of axis orthogonal hyperrectangles into which axis orthogonal systems partition the instance space. In consequence, there is reason to believe that convex hulls might provide a more exible and general learning bias than axis orthogonal regions. The formation of convex hulls around a group of points of the same class is shown to be a usable generalisation and is more general than generalisations produced by axis-orthogonal based classi ers, without constructive induction, like decision trees, decision lists and rules. The use of a small number of large hulls as a concept representation is shown to provide classi cation performance which can be better than that of classi ers which use a large number of small fragmentary regions for each concept. A convex hull based classi er, CH1, has been implemented and tested. CH1 can handle categorical and continuous data. Algorithms for two basic generalisation operations on hulls, in ation and facet deletion, are presented. The two operations are shown to improve the accuracy of the classi er and provide moderate classi cation accuracy over a representative selection of typical, largely or wholly continuous valued machine learning tasks. The classi er exhibits superior performance to well-known axis-orthogonal-based classi ers when presented with domains where the underlying decision surfaces are not axis parallel. The strengths and weaknesses of the system are 1 identi ed. One particular advantage is the ability of the system to model domains with approximately the same numberof structures as there are underlying concepts. This leads to the possibility of extraction of higher level mathematical descriptions of the induced concepts, using the techniques of computational geometry, which is not possible from a multiplicity of small regions.

show abstract

“…It is somehow connected with postulates of a constructive induction to the change of the input data (cf. [2]). …”

Section: Multistrategic Approach To Determining the Relevant Attributesmentioning

confidence: 99%

Rough set theory and rule induction techniques for discovery of attribute dependencies in medical information systems

Stefanowski

Słowiński

1997

Principles of Data Mining and Knowledge Discovery

View full text Add to dashboard Cite

A b s t r a c t . Problems connected with applications of the rough set theory to identify the most important attributes and with induction of decision rules from the medical data set are discussed in this paper. The medical data set concerns patients with multiple injuries. The direct use of the original rough set model leads to finding too many possibilities of reducing the input data. To solve this difficulty, a new approach integrating rough set theory, rule induction and statistical techniques is introduced. First, the Chi-square test is additionally performed in order to reject non-significant attributes. Then, starting from remaining attributes we try to construct such definitions of new attributes that improve finally discovered decision rules. The results have shown that the proposed approach integrating all methods has given better results than those obtained by applying the original rough set method. I n t r o d u c t i o nOne of the essential issues in the analysis of experience represented in data sets is looking for dependencies among attributes describing objects of our interest. It is usually assumed that objects are described by two kinds of attributes: condition (features characterizing object properties by means of numerical and symbolic values) and decision (referring to decision, diagnosis and, in general, expressing classification of objects) attributes.In this paper, we consider the discovery of attribute dependencies as two subtasks:discovery of redundant attributes in the representation of dependencies between condition and decision attributes and identification of the most important condition attributes, -discovery of decision rules characterizing the dependency between values of condition attributes and decision attributes.In recent years the authors tried to solve the above tasks in the analysis of various data sets, mainly medical ones, using several data mining approaches

show abstract

Data-driven constructive induction in AQ17-PRE: A method and experiments

Cited by 17 publications

References 4 publications

Integrating machine learning with knowledge acquisition through direct interaction with domain experts

Integrating machine learning with knowledge acquisition through direct interaction with domain experts

Convex Hulls in Concept Induction

Rough set theory and rule induction techniques for discovery of attribute dependencies in medical information systems

Contact Info

Product

Resources

About