Abstract not found

When Newell introduced the concept of the knowledge level as a useful level of description for computer systems, he focused on the representation of knowledge. This paper applies the knowledge level notion to the problem of knowledge acquisition. Two interesting issues arise. First, some existing machine learning programs appear to be completely static when viewed at the knowledge level. These programs improve their performance without changing their "knowledge." Second, the behavior of some other machine learning programs cannot be predicted or described at the knowledge level. These programs take unjustified inductive leaps. The first programs are called symbol level learning (SLL) programs; the second, non-deductive knowledge level learning (NKLL) programs. The paper analyzes both of these classes of learning programs and speculates on the possibility of developing coherent theories of each. A theory of symbol level learning is sketched, and some reasons are presented for believing...

This article describes how a reasoner can improve its understanding of an incompletely understood domain through the application of what it already knows to novel problems in that domain. Casebased reasoning is the process of using past experiences stored in the reasoner's memory to understand novel situations or solve novel problems. However, this process assumes that past experiences are well understood and provide good "lessons" to be used for future situations. This assumption is usually false when one is learning about a novel domain, since situations encountered previously in this domain might not have been understood completely. Furthermore, the reasoner may not even have a case that adequately deals with the new situation, or may not be able to access the case using existing indices. We present a theory of incremental learning based on the revision of previously existing case knowledge in response to experiences in such situations. The theory has been implemented in a case-base...

interpreter for logic programs (Sterling & Shapiro, 1986)...................138 1 1. INTRODUCTION The most important outcome of AI research during the 70s was the general acceptance of the major role of knowledge in intelligent systems (Buchanan & Feigenbaum, 1982). Lenat and Feigenbaum (1989) call this belief the knowledge as power hypothesis and assert it as: "The knowledge principle (KP) A system exhibits intelligent understanding and action at a high level of competence primarily because of the specific knowledge that it can bring to bear: the concepts, facts, representations, methods, models, metaphors, and heuristics about its domain of endeavor." Or as Buchanan and Feigenbaum (Buchanan & Feigenbaum, 1982) put it, "the power of an intelligent program to perform its task well depends primarily on the quantity and quality of knowledge it has about that task." Thus, it is not surprising that the general attitude toward knowledge was a greedy one - grab as much knowledge as you ca...

Given a sequence of events (or ob]ects), each 'characterized by a set of attributes, the problem considered is to discover a rule characterizing the sequence and able to predict a plausible sequence continuation. The rule, called a sequence-generating rule, is nondeterministic in the sense that it does not necessarily tell exactly which etent must appear next in the sequence, but rather, defines a set of plausible next eents. The basic assumption of the methodology presented here is that the next etent depends solely on the attributes of the previous eents in the sequence. These attributes are either initially given or can be den'td from the initial ones through a chain of inferences. Three basic rule models are employed to guide the search for a sequence.generating rule: decomposition, periodic, and disjunctive normal form (DNF). The search process involves simultaneously transforming the initial sequences to derived sequences and instantiating models to find the best match between the instantiated model and the derived sequence. A program, called SPARC/E, is described that implements most of the methodology a.v applied to discosring sequence generating rules in the card game Eleusis. This game, which models the process of scientiftc discovery, is used as a sottrce of examples for illustrating the performance of SPARC/E.

This thesis describes a machine learning methodology called conjunctive conceptual clustering. The methodology can find conceptual patterns in data as illustrated by three sample problems. In one problem, the method is used to rediscover categories oC soybe3n disc3Se when given a collection oC 4i descriptions oC dise3Sed soybeans h3ving one of Cour diseases. In a second problem. the method is used to find c3tegories underlying a collection of blocks-world structures. In 3 third problem. c3tegories of objects h3ving a more complex structure 3re determined 3nd contr3Stcd with categories gener3ted by people. The described method of conjunctive conceptual clustering Corms clusters oC objects (or situations) not on the b3Sis oC 3 numeric31 similarity me3Sure but on the b3Sis of the "conceptu31 cohesiveness " of one object to 3nother. The conceptu3l cohesiveness between two objects depends on the descriptions oC the two objects 3S well 3S the descriptions oC other ne3rby objects in the given collection 3nd concepts which are 3v3i13ble to describe object groups or object configurations 3S a whole. From a collection oC objects, some b3ckground domain knowledge, and a goal or purpose Cor clustering, conceptual clustering generates a hierarchic3l cl3Ssific3tion

Knowledge Discovery in Databases (KDD) is an active research area with the promise for a high payoff in many business and scientific applications. The grand challenge of knowledge discovery in databases is to automatically process large quantities of raw data, identify the most significant and meaningful patterns, and present this knowledge in an appropriate form for achieving the user's goal. Knowledge discovery systems face challenging problems from the real-world databases which tend to be very large, redundant, noisy and dynamic. Each of these problems has been addressed to some extent within machine learning, but few, if any, systems address them all. Collectively handling these problems while producing useful knowledge efficiently and effectively is the main focus of the thesis. In this thesis, we develop an attribute-oriented rough set approach for knowledge discovery in databases. The method adopts the artificial intelligent "learning from examples" paradigm combined with rough...

This paper describes how a reasoner can improve its understanding of an incompletely understood domain through the application of what it already knows to novel problems in that domain. Recent work in AI has dealt with the issue of using past explanations stored in the reasoner's memory to understand novel situations. However, this process assumes that past explanations are well understood and provide good "lessons" to be used for future situations. This assumption is usually false when one is learning about a novel domain, since situations encountered previously in this domain might not have been understood completely. Instead, it is reasonable to assume that the reasoner would have gaps in its knowledge base. By reasoning about a new situation, the reasoner should be able to fill in these gaps as new information came in, reorganize its explanations in memory, and gradually evolve a better understanding of its domain. We present a story understanding program that retrieves past explan...

We describe the learning system CiPF, which tightly couples a simple concept learner with a sophisticated constructive induction component. It is described in terms of a generic architecture for constructive induction. We focus on the problem of controlling the abundance of opportunities for constructively adding new attributes. In CiPF the so-called Minimum Description Length (MDL) principle acts as a powerful control heuristic. This is also confirmed in the experiments reported. 1 Introduction In learning concept descriptions from preclassified examples, simple concept learners typically make strong assumptions about the way these examples are represented. For effectively learning a concept its examples must populate one or a few regions of the hypothesis space expressible in the description language. For example, decision trees encode axis-parallel nested hyper-rectangles. Two different problems may cause irregular distributions of learning examples in the original representation s...

We describe CIPF 2.0, a propositional constructive learner which is able to cope with both noise and representation mismatch in training examples simultaneously. CIPF 2.0's abilities stem from coupling the robust selective learner C4.5 (and its production rule generator) with a sophisticated constructive induction component. An important new general constructive operator incorporated into CIPF 2.0 is the simplified Kramer operator which abstracts combinations of two attributes into a single new boolean attribute. The so-called Minimum Description Length (MDL) principle acts as a powerful control heuristic guiding the search in the possibly vast representation space. 1 INTRODUCTION When learning concept descriptions from preclassified examples, simple concept learners typically make strong assumptions about the way these examples are represented. For a concept to be learnable, its examples must populate one or a few regions of the hypothesis space expressible in the description languag...