T.-S. Lim, W.-Y. Loh, and Y.-S. Shih, `An empirical comparison of decision trees and other classification methods', Technical Report 979, Madison, WI, (30 1997). Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
RainForest - A Framework for Fast Decision Tree.. - Gehrke, Ramakrishnan.. (1998) (26 citations) (Correct)
....for three reasons. First, due to their intuitive representation, they are easy to assimilate by humans [BFOS84] Second, they can be constructed relatively fast compared to other methods [MAR96, SAM96] Last, the accuracy of decision tree classifiers is comparable or superior to other models [LLS97, Han97] In this paper, we restrict our attention to decision tree classifiers. Within the area of decision tree classification, there exist a large number of algorithms to construct decision trees (also called classification trees; we will use both terms interchangeably) Most algorithms in the ....
....we intend to explore their use in future research. 5 Experimental results In the machine learning and statistics literature, the two main performance measures for classification tree algorithms are: i) The quality of the rules of the resulting tree, and (ii) The decision tree construction time [LLS97] The generic schema described in Section 3 allows the instantiation of most (to our knowledge, all) classification tree algorithms from the literature without modifying the result of the algorithm. Thus, quality is an orthogonal issue in our framework, and we can concentrate solely on decision ....
[Article contains additional citation context not shown here]
T.-S. Lim, W.-Y. Loh, and Y.-S. Shih. An empirical comparison of decision trees and other classification methods. TR 979, Department of Statistics, UW Madison, June 1997.
RainForest - A Framework for Fast Decision Tree.. - Gehrke, Ramakrishnan.. (1998) (26 citations) (Correct)
....for three reasons. First, due to their intuitive representation, they are easy to assimilate by humans [BFOS84] Second, they can be constructed relatively fast compared to other methods [MAR96, SAM96] Last, the accuracy of decision tree classifiers is comparable or superior to other models [LLS97, Han97] In this paper, we restrict our attention to decision tree classifiers. Within the area of decision tree classification, there exist a large number of algorithms to construct decision trees (also called classification trees; we will use both terms interchangeably) Most algorithms in the ....
....we intend to explore their use in future research. 5 Experimental results In the machine learning and statistics literature, the two main performance measures for classification tree algorithms are: i) The quality of the rules of the resulting tree, and (ii) The decision tree construction time [LLS97] The generic schema described in Section 3 allows the instantiation of most (to our knowledge, all) classification tree algorithms from the literature without modifying the result of the algorithm. Thus, quality is an orthogonal issue in our framework, and we can concentrate solely on decision ....
[Article contains additional citation context not shown here]
T.-S. Lim, W.-Y. Loh, and Y.-S. Shih. An empirical comparison of decision trees and other classification methods. TR 979, Department of Statistics, UW Madison, June 1997.
A Survey of Methods for Scaling Up Inductive Algorithms - Provost, Kolluri (1999) (31 citations) (Correct)
....founded algorithm, called T2, for learning two level decision trees. They show that for eight out of fifteen data sets, T2 produces two level trees which rival or surpass the de facto standard C4.5 (see below) Quinlan 1993) Interestingly, in practice C4.5 is considerably faster than T2 (Lim, Loh, and Shih 1999) more on this presently. 6.1.2. Powerful Search Heuristics Certainly, in some domains there is leverage to be gained by searching for more complex models. The size and structure of the space of models, the size of the sample necessary to learn well, and the computational 8 PROVOST AND KOLLURI ....
....go on to show that similar techniques can be used to speed up learning with hierarchically structured data (Almuallim, Akiba, and Kaneda 1995) to which we will return when we discuss relational representations. Among the various decision tree programs, C4.5 has been shown to be comparably fast (Lim, Loh, and Shih 1999) remarkably so considering the programs similarity. An analysis of its code shows that when evaluating node splits, C4.5 first builds a sufficient statistics contingency table, and then uses it to decide on the best split. Kufrin (1997) notes that preliminary experiments with additional ....
Lim, T.-J., W.-Y. Loh, and Y.-S. Shih (1999). An empirical comparison of decision trees and other classification methods. Machine Learning. To appear.
An Empirical Evaluation of Supervised Learning - For Roc Area (Correct)
No context found.
T.-S. Lim, W.-Y. Loh, and Y.-S. Shih, `An empirical comparison of decision trees and other classification methods', Technical Report 979, Madison, WI, (30 1997).
An Empirical Evaluation of Supervised Learning for ROC Area - Caruana, Niculescu-Mizil (2004) (Correct)
No context found.
T.-S. Lim, W.-Y. Loh, and Y.-S. Shih, `An empirical comparison of decision trees and other classification methods', Technical Report 979, Madison, WI, (30 1997).
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