Quinlan, J. R. (1983). Learning efficient classification procedures and their application to chess end games. In Michalski, Carbonell & Mitchell (Eds.), Machine learning: An artificial intelligence approach. San Mateo, CA: Morgan Kaufmann.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....algorithms are presented in Chapter 3 in more detail (Section 3.1 and 3.2) 2.2 Decision Trees Decision trees are one of the most well known and widely used approaches for learning from examples. This method was developed initially by Hunt, Marin and Stone [31] and later modified by Quinlan [49, 50]. Quinlan s ID3 [52] and C4.5 [55] are the most popular algorithms in decision tree induction. Initially, ID3 algorithm has applied to deterministic domains such as chess and games [49, 50] Later, ID3 algorithm has extended to cope with noisy and uncertain instances rather than being ....

....examples. This method was developed initially by Hunt, Marin and Stone [31] and later modified by Quinlan [49, 50] Quinlan s ID3 [52] and C4.5 [55] are the most popular algorithms in decision tree induction. Initially, ID3 algorithm has applied to deterministic domains such as chess and games [49, 50]. Later, ID3 algorithm has extended to cope with noisy and uncertain instances rather than being deterministic [52] Decision tree algorithms represents concept descriptions in the form of tree structure. Decision tree algorithms begin with a set of instances and create a tree data structure that ....

J.R. Quinlan, Learning Efficient Classification Procedures and Their Application to Chess and Games, In R.S. Michalski, J.G. Carbonell, and T.M. Mitchell (Eds.), Machine Learning, An Artificial Intelligence Approach, Los Altos: Morgan Kaufmann, 1983.

....concept and the subset not covered. Any mechanism which produces such blackboxes can be construed as a concept learner. The class of all such mechanisms is large. It encompasses concept learners proper, e.g. symbolist methods such as Candidate Elimination [1, 2] Focussing [3,4] Classification [5] and Conceptual Clustering [6,Fisher, Learning from 7,8,9] It also encompasses mechanisms which do not have concept learning as an explicit goal but which nevertheless produce the requisite black boxes, e.g. connectionist mechanisms such as BackPropagation, Competitive Learning [10] and ....

....levels of output for certain classes of input so can be thought of as probabilistic concepts covering the classes in question. Clustering mechanisms produce dendrograms which can be thought of as disjunctive concepts in the manner of the decision trees produced by symbolist mechanisms such as ID3 [5,13]. The fact that so many computational mechanisms do (or can be seen as doing) concept learning seems to suggest that the process must be rather significant cognitively speaking. But what is its significance Putting the question another way: What is the point of concept learning Why is it a good ....

Quinlan, J. (1983). Learning efficient classification procedures and their application to chess end games. In R. Michalski, J. Carbonell and T. Mitchell (Eds.), Machine Learning: An Artificial Intelligence Approach. Palo Alto: Tioga.

....hyperplane into position. Over time, LMS regression has come to be seen as a major foundation for classification methods applied to numerical data [12] Data mining also has roots outside of statistics. The decision tree method, for example, has a history which embraces Quinlan s ID3 method [13, 14], a program developed within the machine learning (artificial intelligence) community. Learning here involves producing a decision tree for predictions which effectively minimises the number of tests which have to be made in order to generate a prediction consistent with the training examples. ....

Quinlan, J. (1983). Learning efficient classification procedures and their application to chess end games. In R. Michalski, J. Carbonell and T. Mitchell (Eds.), Machine Learning: An Artificial Intelligence Approach. Palo Alto: Tioga.

....context sensitive measure might involve further extensions of the PAC learning model. However, the present paper explores the possibility of deriving a measure from information theory, which is similar to the method for computing the information deficit of decision tree nodes (as used in, e.g. [11,12]) It shows how we can modify the method to produce a measure of hypothesis uncertainty and then derive a learner uncertainty measure as the mean of the uncertainties for the relevant hypothesis set. This uncertainty measure forms an inverse measure of the effectiveness of the learner s ....

....selected attributes. The information theoretic heuristic is applied at the point where an attribute must be selected on which to split the instances at a given leaf node. The essential idea is that an ideal split should create a set of child nodes whose average information deficit is minimized [11]. The information deficit of node X is simply the amount of information required to produce a true classification of an arbitrary instance that currently classifies at X. To compute this, we look at the relative frequencies of positive and negative instances at X and derive probabilities for the ....

[Article contains additional citation context not shown here]

Quinlan, J. (1983). Learning efficient classification procedures and their application to chess end games. In R. Michalski, J. Carbonell and T. Mitchell (Eds.), Machine Learning: An Artificial Intelligence Approach. Palo Alto: Tioga.

.... to describe the training and test instances (Almuallim and Dietterich, 1991, Langley and Sage, in press) Unfortunately, the task of designing an appropriate instance representation also known as feature set selection can be extraordinarily difficult, time consuming, and knowledge intensive (Quinlan, 1983). This poses a problem for current statistical and machine learn ing approaches to natural language understanding where a new instance representation is typically required for each linguistic task tackled. This paper addresses the role of the underlying instance representation for one class of ....

J. R. Quinlan. 1983. Learning Ef- ficient Classification Procedures and Their Application to Chess End Games. In R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, editors, Machine Learning: An Artificial Intelli- 126 gence Approach. Morgan Kaufmann, San Mateo, CA.

....f (net) where f (net) is the sigmoid function. EEwoo jjkjjj k = 17 where k ranges over all nodes in the layers above node j. The process will repeat for each iteration of computation. 4. Designing the neural network application in auditing Recently, a number of researchers [16 19] have proposed using machine learning techniques for modeling expert knowledge. Neural networks represent an approach to this issue. Unlike traditional expert systems, where knowledge is made explicit in the form of rules, neural networks generate their own rules by training examples. Learning is ....

....in predicting. Two further researches are needed. First, how does one know how large a training set is sufficient In general, this question is in need of research, although recent work [20] has resulted in a theory of lower bounds on the number of examples required for learning. Quinlan [19] has done some preliminary work on estimating the necessary size of the training set. Still, currently there is no established theory that applies. Second, there is a need for better tools for estimating convergence time. Some neural networks, back propagation included, cannot be guaranteed to ....

J.R. Quinlan, Learning efficient classification procedures and their application to chess end games in: Machine Learning: An Artificial Intelligence Approach, R.S. Michalski, J.G. Carbonell and T.M. Mitchel, eds., Tioga, Palo Alto, CA, 1983.

....because of the substantial efforts for construction and maintenance of concept hierarchies in large databases. There have been many interesting studies on automatic generation of concept hierarchies for nominal data, which can be categorized into different approaches: machine learning approaches [15, 4, 19], statistical approaches [1] visual feedback approaches [12] algebraic (lattice) approaches [16] etc. Machine learning approach for concept hierarchy generation is a problem closely related to concept formation. Many influential studies have been performed on it, including Cluster 2 by ....

....[12] algebraic (lattice) approaches [16] etc. Machine learning approach for concept hierarchy generation is a problem closely related to concept formation. Many influential studies have been performed on it, including Cluster 2 by Michalski and Stepp [15] COBWEB by Fisher [4] ID3 by Quinlan [19], hierarchical and parallel clustering by long and Mao [10] and many others. These approaches are under our careful examination and experimentation and our goal is to develop an efficient algorithm to maximize the automatic data clustering capability for large databases. Our progress will be ....

J. R. Quinlan. Learning efficient classification procedures and their application to chess end-games. In Michalski et. al., editor, Machine Learning: An Artificial Intelligence Approach, Vol. 1, pages 463-482. Morgan Kaufmann, 1983.

....# L. More specifically, # is a description of m(#) in S, the intension of concept (#, m(#) and m(#) is the set of objects satisfying #, the extension of concept (#, m(#) To illustrate the idea developed so far, consider an information table given by Table 1, which is adopted from Quinlan [5]. The following expressions are some of the formulas of the language L: height = tall, hair = dark. The meanings of the formulas are given by: m(height = tall) m(hair = dark) By pairing intensions and extensions, we can obtain formal concepts such as (height = ....

....definable partition lattice introduced earlier. For example, by search the semilattice # CD(C) U) we can obtain classification rules whose left hand sides are only conjunction of atomic formulas. The well known ID3 learning algorithm in fact searches # CD(C) U) for classification rules [5]. By searching the lattice # AD(C) U) one can obtain a similar solution. We can re express many fundamental notions of classification in terms of partitions. Definition 11 For two solutions # 1 , # 2 ## of a consistent classification problem, namely, # 1 # class , if # 1 # 2 , we ....

Quinlan, J.R. Learning efficient classification procedures and their application to chess end-games, in: Machine Learning: An Artificial Intelligence Approach, Vol. 1, Michalski, J.S., Carbonell, J.G., and Mirchell, T.M. (Eds.), Morgan Kaufmann, Palo Alto, CA, pp. 463-482, 1983.

....feature of this approach is that the knowledge on which the active vision is based is learned, and not preprogrammed, as in many other approaches. Also, as we will show, the method performs quite well in the presence of noise, which is not the case for other feature selection methods such as ID3 [8]. 1Real World Computing Partnership 2Foundation for Neural Networks 2 Boltzmann Machines Boltzmann Machines are stochastic networks. The neurons can be in two states (ri = q l, but also continuous neuron values are possible [5, 7] Using Glauber dynamics, neurons are randomly selected and ....

J.R. Quinlan. Learning efficient classification procedures. In R.S. Michalski, J.G. Carbonell, and T.M. Mitchell, editors, Machine Learning: an artificial intelligence appraoch, pages 463-482. Palo Alto: Tioga, 1983.

....special purpose hardware. This model has been so successful that little else has been tried. The alternative AI approaches have not fared well due to the expense in applying the knowledge that had been supplied to the system. Those times in recent years that chess has been applied as a testbed [8, 27, 19, 21, 22, 30, 33, 26, 20] only a small sub domain of the game was used, so that fundamental efficiency issues that AI must grapple with have been largely unaddressed. However, we feel that there is a third approach that neither relies on search or the symbolic computation approach of knowledge oriented AI: what we shall ....

J. R. Quinlan. Learning efficient classification procedures and their application to chess end games. In R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, editors, Machine Learning. Morgan Kaufmann, San Mateo, CA, 1983. 12

....form of the assumption [Minsky and Papert, 1969; Jordan and Jacobs, 1990] Unless chosen with fore knowledge of the structure of the world model, there is no guarantee that any possible set of parameters could produce a mapping which would adequately model the data. Decision tree 20 classifiers [Quinlan, 1983] make a weaker assumption that the domain can be split up into a fairly small number of large hyperrectangular regions in which the classification is constant. This feature is common with the approach of [Salzberg, 1988] and [Aha et al. 1990] which both learn using the nearest neighbout ....

J. R. Quinlan. Learning Efficient Classification Procedures and their Application to Chess End Games. In R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, editors, Machine Learning--An Artificial Intelligence Approach (I). Tioga Publishing Company, Palo Alto, 1983.

....accuracy, we propose a weighted sampling, based on a boosting technique [4] where the prediction models in subsequent iterations are built on those examples on which the previous predictor had poor performance. Similar techniques of active or controllable sampling are related to windowing [5], wherein subsequent sampling chooses training instances for which the current model makes the largest errors. However, simple active sampling is notoriously ill behaved on noisy data, since subsequent samples contain increasing amount of noise and performance often decrease as sampling progresses ....

Quinlan, J. R.: Learning Efficient Classification Procedures and their Application to Chess and Games, In Michalski, R., Carbonell, J., Mitchell, T. (eds.): Machine Learning. An Artificial Intelligence Approach, (1983), 463-482

No context found.

Quinlan, J. R. (1983). Learning efficient classification procedures and their application to chess end games. In Michalski, Carbonell & Mitchell (Eds.), Machine learning: An artificial intelligence approach. San Mateo, CA: Morgan Kaufmann.

No context found.

Quinlan, J.R.: "Learning Efficient Classification Procedures and their Application to Chess End Games", in [ML-Vol I], 463-482.

No context found.

Quinlan, J. R. (1983). Learning efficient classification procedures and their application to chess end games (pp. 463-482). In Michalski, Carbonell & Mitchell (Eds.), Machine learning: An artificial intelligence approach. San Mateo, CA: Morgan Kaufmann.

No context found.

QUINLAN, J. R., Learning efficient classification procedures and their application to chess end games. In R. S. MICHALSKI, J. G. CARBONELL & T. MITCHELL (Eds.), Machine learning: An artificial intelligence approach, Palo Alto, Tioga (1983), 463-482.

No context found.

QUINLAN, J. R., Learning efficient classification procedures and their application to chess end games. In R. S. MICHALSKI, J. G. CARBONELL & T. MITCHELL (Eds.), Machine learning: An artificial intelligence approach, Palo Alto, Tioga (1983), 463-482.

No context found.

Quinlan, J.R.: "Learning Efficient Classification Procedures and their Application to Chess End Games", in [ML-Vol I], 463-482.

No context found.

Quinlan J. R. Learning efficient classification procedures and their application to chess end games. R. S. Michalski, J. G. Carbonell and T. M. Mitchell (Eds.), Machine learning: An artificial intelligence approach. San Mateo, CA: Morgan Kaufmann, 1983.

No context found.

J. R. Quinlan. Learning efficient classification procedures and their application to chess end games. In R. S. Michalski, J.G. Carbonelli, and T. M. Mitchell, editors, Machine Learning -- An Artificial Intelligenc Approach, pages 463--482. Springer-Verlag, 1983.

No context found.

J. R. Quinlan. Learning Efficient Classification Procedures and their Application to Chess End Games. In R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, editors, Machine Learning--- An Artificial Intelligence Approach (I). Tioga Publishing Company, Palo Alto, 1983.

No context found.

Quinlan J. R. Learning efficient classification procedures and their application to chess end games. R. S. Michalski, J. G. Carbonell and T. M. Mitchell (Eds.), Machine learning: An artificial intelligence approach. San Mateo, CA: Morgan Kaufmann, 1983.

No context found.

Quinlan, J. (1983). Learning efficient classification procedures and their application to chess end games. In R. Michalski, J. Carbonell and T. Mitchell (Eds.), Machine Learning: An Artificial Intelligence Approach. Palo Alto: Tioga.

No context found.

Quinlan, J. (1983). Learning efficient classification procedures and their application to chess end games. In R. Michalski, J. Carbonell and T. Mitchell (Eds.), Machine Learning: An Artificial Intelligence Approach. Palo Alto: Tioga.

No context found.

R. Quinlan. Learning efficient classification procedures and their application to chess end games. In Ryszard Michalski, Jaime Carbonell, and Tom Mitchell, editors, Machine Learning, pages 463--482. Morgan Kaufman, 1983.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC