Murthy, S.K.: Automatic Construction of Decision Trees from Data: A MultiDisciplinary Survey. Data Mining and Knowledge Discovery, 2 4 (1998) 345-389  Home/Search   Document Details and Download   Summary   Related Articles   Check

....are identified. 1 Introduction Classification problems form an important area in datamining. For example, a bank may want to classify its clients in good and bad credit risks or a doctor may want to classify his patients as having diabetes or not. Classifiers may take the form of decision trees [11] (see Figure 1) In each node, a test is made in which one or more variables is used. Depending on the outcome of the test, the tree is traversed to the left or the right subtree (see Section 2.1) In our decision trees, the tests are linear combinations of some of the variables. This allows ....

....valued datasets with an (unknown) inherent linear structure. An optimal tree is one which makes as few misclassifications as possible on the validation set. Well known decision tree algorithms such as ID3, CART, OC1 and C4.5 are greedy local search algorithms which construct trees top down [11]. In this paper, genetic programming (GP) 5] is used as a global stochastic search technique for finding accurate decision trees. Previous work on evolving decision trees with GP was done in [5] and [12] The standard representation of GP was used in these experiments. In this paper, a new ....

[Article contains additional citation context not shown here]

S. K. Murthy. Automatic construction of decision trees from data: a multidisciplinary survey. In Data Mining and Knowledge Discovery, number 2, pages 345--389, 1998.

....it impractical. One way to solve this problem is to use a decision tree. A well designed decision tree can retrieve a matched sample with a logarithmic time complexity. This is a very useful characteristic for large image databases. There is a very rich literature about decision trees, see surveys [8] [3] The applications of decision tree have been traditionally for a low dimensional feature space with manually selected features. This is true largely because humans cannot define a large number of useful features. Appearance based approach drastically changed this situation. Traditional ....

S. K. Murthy. Automatic construction of decision trees from data: A multidisciplinary survey. Data Mining and Knowledge Discovery, 1998.

....a linear regression tree algorithm. We show results of an extensive experimental study of SECRET in Section 5, and we conclude in Section 6. 2. PRELIMINARIES In this section we give a short introduction to classification and regression trees. More details can be found in the excellent review [13]. 2.1 Classification Trees Classification trees have a tree organization with intermediate nodes labeled by split attributes, the branches starting in an intermediate node labeled by split predicates involving the corresponding split predicate and leaves labeled by class labels. Prediction ....

....proceeds in two phases. In the growing phase an oversized tree is build recursively, at every step a split attribute and split predicates involving this attribute are selected in order to maximize the goodness of split criteria. A large number of such criteria have been proposed in the literature [13], here we use Breiman s gini gain since is easy to compute and the best split for discrete attributes can be found e#ciently [3] In the second phase the final size of the tree is determined with the goal to minimize the error on unseen examples. 2.2 Regression Trees Regression models or ....

S. K. Murthy. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and Knowledge Discovery, 1997.

....learn how to recognize the type of the workload mix. The workload itself contains valuable information about its characteristics; this information that can be extracted and analyzed. Our approach is, therefore, to use data mining classification techniques, specifically Decision Trees Induction [17], to build a classification model. One of the advantages of using decision tree induction is its high interpretability, that is, the ease of extracting the classification rules and the ability to understand and justify the results, in comparison with other techniques such as neural networks. 3.1 ....

S. Murthy. Automatic Construction of Decision Trees from Data: A Multi- disciplinary Survey. Data Mining and Knowledge Discovery 2, pages 345--389, 1998.

....trees have proven to be a valuable tool for description, classification and generalization of data. This is related to the compact and intelligible representation of the learned classification function, and the availability of a large number of efficient algorithms for their automated construction [8]. They provide a hierarchical way to represent rules underlying data. Today, a wealth of algorithms for the automatic construction of decision trees can be traced back to the ancestors ID3 [9] or CART [3] In many practical applications, the data used are inherently of imprecise and subjective ....

S. K. Murthy. Automatic construction of decision trees from data: a multi-disciplinary survey. Data Mining and Knowledge Discovery, 2:345--389, 1998.

....in the training data. Parameters can also be tied for the covariance matrix of the distribution using algorithms such as semi tied covariance [40, 11, 46, 41] Decision trees have proved to be useful in classi cation, description and generalization of data in a variety of applications (e.g. [91, 14, 15]) Decision trees were rst used in ASR for tying HMMs where state labels were approximated using a Poisson process [5] It became popular when they were used to cluster observation densities of a state in a maximum likelihood framework [70, 130] Decision tree clustering is particularly ....

....also increases with number of features. A variety of methods have been attempted to address the problem of data sparsity, including use of decision graphs [76, 96] pylons [4] and soft decisions [107] for additional pointers, see Murthy s comprehensive multi disciplinary survey on decision trees [91]) however, no clear solution has emerged. 5.2 Multi stage Clustering Our approach to reduce the storage and computational costs for clustering is based on dividing the task into multiple stages. The decision tree can be viewed as a function, T , that maps a feature vector, f , consisting of ....

Sreerama K. Murthy. Automatic construction of decision trees from data: a multidisciplinary survey. Data Mining and Knowledge Discovery, 2(4):345-389, 1998.

....the data hitting node m, starting with the complete dataset in deciding on the root node. Once we decide on a split, tree construction continues recursively for each child with training instances taking that branch. Surveys about constructing and simplifying decision trees can be found in [6] and [15]. A recent survey comparing different decision tree methods with other classification algorithms is given in [11] The best split is when all the instances from a class lie on the same side of the decision boundary, i.e. return the same truth value for fm . There are various measures proposed ....

S. K. Murthy, "Automatic construction of decision trees from data: A multidisciplinary survey.," Data Mining and Knowledge Discovery, vol. 4, pp. 345--389, 1998.

....on clusters of SMPs. Our middleware is based on the observation that parallel versions of several well known data mining techniques, including apriori association mining [1] k means clustering [14] knearest neighbor classifier [13] artificial neural networks [13] and decision tree classifiers [17] share a relatively similar structure. The middleware performs distributed memory parallelization across the cluster and shared memory parallelization within each node. It enables high I O performance by minimizing disk seek time and using asynchronous I O operations. Thus, it can be used for ....

....memory machine in a very similar way. Our discussion focuses on three important techniques: apriori associating mining [1] k means clustering [14] and k nearest neighbors [13] The same argument applies for artificial neural networks [13] bayesian networks [8] and decision tree classifiers [17], and will be presented in the final paper. 2.1 Apriori Association Mining Association rule mining is the process of analyzing a set of transactions to extract association rules and is a very commonly used and well studied data mining problem [2, 29] Given a set of transactions 1 (each of ....

S. K. Murthy. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and Knowledge Discovery, 2(4):345--389, 1998.

No context found.

Murthy, S.K.: Automatic Construction of Decision Trees from Data: A MultiDisciplinary Survey. Data Mining and Knowledge Discovery, 2 4 (1998) 345-389

No context found.

S.K. Murthy, `Automatic construction of decision trees from data: A multi-disciplinary survey', Data Mining and Knowledge Discovery, 2(4), 345--389, (1998).

No context found.

S. K. Murthy. Automatic construction of decision trees from data: a multidisciplinary survey. In Data Mining and Knowledge Discovery, number 2, pages 345--389, 1998.

No context found.

Murthy, S. K. 1998. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and Knowledge Discovery 2, 4, 345--389.

No context found.

Murthy (1998), Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey, Data Mining and Knowledge Discovery, 2, 345--389 (1998), Kluwer Academic Publishers.

No context found.

Murthy S.: Automatic Construction of Decision Trees from Data: A MultiDisciplinary Survey. Data Mining and Knowledge Discovery, Vol. 2, (1998), 345-- 389.

No context found.

S.K. Murthy, "Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey," Data Mining and Knowledge Discovery, vol. 2, no. 4, pp. 345-389, 1998.

No context found.

Murthy (1998), Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey, Data Mining and Knowledge Discovery, 2, 345--389 (1998), Kluwer Academic Publishers.

No context found.

S.K. Murthy. Automatic Construction of Decision Trees from Data: A MultiDisciplinary Survey. Data Mining and Knowledge Discovery, 2(4): 345-389, 1998.

No context found.

S. K. Murthy. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and Knowledge Discovery, 2(4):345--389, 1998.

No context found.

S.K. Murthy. Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey. Data Mining and Knowledge Discovery, 2(4): 345-389, 1998.

No context found.

S. K. Murthy. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and Knowledge Discovery, 2(4):345--389, 1998.

No context found.

S. K. Murthy. Automatic construction of decision trees from data: a multi-disciplinary survey. In Data Mining and Knowledge Discovery, number 2, pages 345--389, 1998.

No context found.

S.K. Murthy. Automatic Construction of Decision Trees from Data: A MultiDisciplinary Survey. Data Mining and Knowledge Discovery, 2(4): 345-389, 1998.

No context found.

S. K. Murthy. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and Knowledge Discovery, 2(4):345-389, 1998.

No context found.

S.K. Murthy. Automatic Construction of Decision Trees from Data: A MultiDisciplinary Survey. Data Mining and Knowledge Discovery, 2(4): 345-389, 1998.

No context found.

Murthy, S.K. (1998). Automatic construction of decision trees from data: a multidisciplinary survey. Data Mining and Knowledge Discovery 2, 345--389.

First 50 documents

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