Modifying knowledge-based systems is a complex activity. One of its di#culties is that several related portions of the system mighthavetobechanged in order to maintain the coherence of the system. However, it is di#cult for users to #gure out what has to be changed and how. This paper presents a novel approach for building knowledge acquisition tools that overcomes some of the limitations of current approaches. In this approach, knowledge of prototypical procedures for modifying knowledge-based systems is used to guide users in changing all related portions of a system. These procedures, whichwe call knowledge acquisition scripts #or KA Scripts#, capture how related portions of a knowledge-based system can be changed coordinately.By using KA scripts, a knowledge acquisition tool would be able to relate individual changes in di#erent parts of a system, enabling the analysis of each individual change from the perspective of the overall modi#cation. The paper also describes the ...

Rapid prototyping and tool reusability have pushed knowledge acquisition research to investigate method-specific knowledge acquisition tools appropriate for predetermined problem-solving methods. We believe that method-dependent knowledge acquisition is not the only approach. The aim of our research istodevelop powerful yet versatile machine learning mechanisms that can be incorporated into general-purpose but practical knowledge acquisition tools. This paper shows through examples the practical advantages of this approach. In particular, we illustrate how existing knowledge can be used to facilitate knowledge acquisition through analogy mechanisms within a domain and across domains. Our sample knowledge acquisition dialogues with a domain expert illustrate which parts of the process are addressed by the human and which parts are automated by the tool, in a synergistic cooperation for knowledge-base extension and re nement. The paper also describes briefly the expect problem-solving architecture that facilitates this approach toknowledge acquisition.

This paper presents a method for constructive induction in which new problem-relevant atu'ibutes are generated by analyzing iteratively created inductive hypotheses. The method starts by creating a set of rules from given examples using the AQ algorithm. These rules are then evaluated according to a "rule quality criterion." Subsets of the best-performing rules for each decision class are selected to form new attributes. These new attributes are used to reformulate the training examples used in the previous step, and the whole inductive process starts again. This iterative process ends when the performance of the rules exceeds a determined threshold. In the experiments on learning different DNF functions, the method outperformed in terms of predictive accuracy both, the AQ15 role learning method, as well as the REEDWOOD decision tree learning method.

This paper is concerned with the task of incrementally acquiring and refining the knowledge and algorithms of a knowledge-based system in order to improve its performance over time. In particular, we present the design of DE-KART, a tool whose goal is to provide increasing levels of assistance in acquiring and refining indexing and retrieval knowledge for a knowledge-based retrieval system. DE-KART starts with knowledge that has been entered manually, and increases its level of assistance in acquiring and refining that knowledge, both in terms of the increased level of automation in interacting with users, and in terms of the increased generality of the knowledge. DE-KART is at the intersection of machine learning and knowledge acquisition: it is a first step towards a system which moves along a continuum from interactive knowledge acquisition to increasingly automated machine learning as it acquires more knowledge and experience.

The ability to build intelligent agents is significantly constrained by the knowledge acquisition effort required. Many iterations by human experts and knowledge engineers are currently necessary to develop knowledge-based agents with acceptable performance. We have developed a novel approach, called Disciple, for building intelligent agents that relies on an interactive tutoring paradigm, rather than the traditional knowledge engineering paradigm. In the Disciple approach, an expert teaches an agent through five basic types of interactions. Such rich interaction is rare among machine learning systems, but is necessary to develop more powerful systems. These interactions, from the point of view of the expert, include: specifying knowledge to the agent; giving the agent a concrete problem and its solution that the agent is to learn a general rule for; validating analogical problems and solutions proposed by the agent; explaining to the agent reasons for the validation; and being guide...

This chapter describes a general framework for multistrategy learning. One idea of this framework is to view learning as an inference process and to integrate the elementary inferences that are employed by the single-strategy learning methods. Another idea is to base learning on building and generalizing a special type of explanation structure called plausible justification tree which is composed of different types of inference and relates the learner's knowledge to the input. In this framework, learning consists of extending and/or improving the knowledge base of the system so that to explain the input received from an external source of information. The framework is illustrated with a specific method that integrates deeply and dynamically explanation-based learning, determination-based analogy, empirical induction, constructive induction, and abduction. 1

. This paper presents several industrial applications of ML in the context of their effort to solve the "KAML problem", i.e., the problem of merging knowledge acquisition and machine learning techniques. Case-based reasoning is a possible alternative to the problem of acquiring highly compiled expert knowledge, but it raises also many new problems that must be solved before really efficient implementations are available. 1 Introduction There are many sides to the description of what an industrial application is. In a recent paper (Kodratoff, Graner, and Moustakis, 1994) we summarized some of the experience gained during the CEC project MLT in counseling a user on which of the many types of machine learning (ML) to use for his special application. In this presentation, we shall consider two of the main subfields of the ones that need merging for an industrial application, seemingly the richest in generating future research problems: validation of KBS, and merging of ML into a kn...

This paper presents the knowledge base refmement rnetl1cx \ of NeoDisciple. The method is based 00 two levels of cooperation. Internally. different learning strategies cooperate in solving the knowledge base refinement problem. Externally. NeoDisciple and the human expen cooperate in solving the problems that are intrinsically difficult for an autonomous learning system as. for instance. the new terms problem and the blame assignment problem. The goal is to show the adequacy of such an approach for me automation ofknowledge acquisition. 1

Abstract not found

This paper presents a learning-based representation of knowledge which is at the basis of the family of Disciple learning agents. It introduces a representation for concepts, generalization and specialization rules, different types of generalizations and specializations, and the representation of the main elements of a knowledge base, including partially learned concepts, problems, and rules. Finally, it provides a formal definition of generalization based on substitutions.