Tree classifier design with a permutation statistic

Li, X; Dubes, C

doi:10.1016/0031-3203(86)90013-0

Cited by 30 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, using the same measure for both test selection and pre-pruning is problematic when it is local to a node because the absolute value of local measures typically vary with sample size (Gleser and Collen 1972;Rounds 1980;Quinlan 1986). Rather than using a local test selection measure that is insensitive to the number of cases in a node (Li and Dubes 1986;Zhou and Dillon 1991;Kalkanis 1993) or adopting a global measure (Sethi and Sarvarayudu 1982), researchers solved this problem by using stopping criteria that use a di erent measure than that used for selecting tests. For example, Quinlan (1986) used the 2 test for stochastic independence as the stopping criterion in ID3, while using an information-theoretic measure to evaluate and select tests in eval().…”

Section: Pre-pruningmentioning

confidence: 99%

Simplifying decision trees: A survey

Breslow

Aha

1997

The Knowledge Engineering Review

243

133

View full text Add to dashboard Cite

Induced decision trees are an extensively-researched solution to classification tasks. For many practical tasks, the trees produced by tree-generation algorithms are not comprehensible to users due to their size and complexity. Although many tree induction algorithms have been shown to produce simpler, more comprehensible trees (or data structures derived from trees) with good classification accuracy, tree simplification has usually been of secondary concern relative to accuracy, and no attempt has been made to survey the literature from the perspective of simplification. We present a framework that organizes the approaches to tree simplification and summarize and critique the approaches within this framework. The purpose of this survey is to provide researchers and practitioners with a concise overview of tree-simplification approaches and insight into their relative capabilities. In our final discussion, we briefly describe some empirical findings and discuss the application of tree induction algorithms to case retrieval in case-based reasoning systems.

show abstract

Section: Pre-pruningmentioning

confidence: 99%

Simplifying decision trees: A survey

Breslow

Aha

1997

The Knowledge Engineering Review

243

133

View full text Add to dashboard Cite

show abstract

“…Trees built by minimizing such criteria have been thoroughly studied, see, Breiman, Friedman, Olshen, and Stone [50], Burshtein, Della Pietra, Kanevsky, and Nádas [55], Chou [58], Ciampi [59], Gelfand and Delp [122], Gelfand, Ravishankar, and Delp [123], [124], Goodman and Smyth [129], Guo and Gelfand [135], Li and Dubes [179], Michel-Briand and Milhaud [196], Quinlan [228], Sethi [248], Sethi and Sarvarayudu [249], Talmon [270], Wang and Suen [292], among others.…”

Section: Splitting Criteriamentioning

confidence: 99%

Learning pattern classification-a survey

Kulkarni

Lugosi

Venkatesh

1998

IEEE Trans. Inform. Theory

145

View full text Add to dashboard Cite

Classical and recent results in statistical pattern recognition and learning theory are reviewed in a two-class pattern classification setting. This basic model best illustrates intuition and analysis techniques while still containing the essential features and serving as a prototype for many applications. Topics discussed include nearest neighbor, kernel, and histogram methods, Vapnik-Chervonenkis theory, and neural networks. The presentation and the large (thogh nonexhaustive) list of references is geared to provide a useful overview of this field for both specialists and nonspecialists.

show abstract

“…These direct approaches can be considered as generalization of decision table conversion approaches with all the available pattern vectors for the design forming the decision table entries. Examples of these direct tree design approaches can be found in [6,9,12,15,18]. While some of these tree SIXTH ASAT CONFERENCE 2 -4 MAY 1995, CAIRO…”

Section: Mutual Information Based Self -Configurationmentioning

confidence: 99%

Neural Network Implementation of Binary Trees

Farag

Ibrahim

1995

International Conference on Aerospace Sciences and Aviation Tec

View full text Add to dashboard Cite

Multiple layer artificial neural network (ANN) structure is capable of implementing arbitrary input-output mappings. Similarly, hierarchical classifiers, more commonly known as decision trees, possess the capabilities of generating arbitrarily complex decision boundaries in an ndimensional space. Given a decision tree, it is possible to restructure it as a multilayered neural network. The objective of this paper is to show how this mapping of decision trees into multilayer neural network structure can be exploited for the systematic design of a class of layered neural networks, called entropy nets, that have far fewer connections.

show abstract

Tree classifier design with a permutation statistic

Cited by 30 publications

References 15 publications

Simplifying decision trees: A survey

Simplifying decision trees: A survey

Learning pattern classification-a survey

Neural Network Implementation of Binary Trees

Contact Info

Product

Resources

About