P.M. Murphy and D.W. Aha (1994). UCI Repository of machine learning databases [http://www.ics.uci.edu/mlearn/mlrepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....(NB) classifier and boosted C4.5 [13, 34] Our version of boosted C4.5 is implemented and provided by Z. Zheng [39, 40] We used 20 datasets in our experiments. Five (5) out of the 20 are our real life application datasets. The rest (12) of them are obtained from UCI Machine Learning Repository [28]. We could not use many datasets in UCI Repository in our evaluation because in these datasets the class distributions are not imbalanced which is the main characteristic of the applications that we are interested in. Thus, only those datasets with imbalanced class distributions are selected in ....

Merz, C. J, and Murphy, P. UCI repository of machine learning databases [http://www.cs.uci.edu/-mleam/MLRepository.html], 1996.

....f and correctly labels any feature vector drawn from the same source as the training set. It is common to call the body of knowledge that classifies data with the label y as the concept of class y. 14 As an example, Table 2. 1 shows a medical database of patients examined for the thyroid disease [ Merz Murphy, 1996 ] In this table, rows x 1 , x 4 represent patients and columns i1 , x i6 and y correspond to their medical profiles (ID, age, sex, test results, medication administered) and their diagnoses (i.e. y 1 corresponds to normal, y 2 to hypothyroidism, and y 3 as hyperthyroidism) ....

....level classifier, the meta classifier, by learning over the meta level training set. This meta level training set is composed by using the base classifiers predictions on the validation set as attribute values, and the true class as the target. An example of such a set on the thyroid disease [ Merz Murphy, 1996 ] with three base classifiers is shown in Table 2.2. The first three columns correspond to the predictions (prognosis) of the base classifiers on four patients from a validation set, while the last column represents the correct class (diagnosis) The aim of this strategy is to correlate the ....

[Article contains additional citation context not shown here]

Merz, C., and Murphy, P. 1996. UCI repository of machine learning databases [http://www.ics.uci.edu/#mlearn/mlrepository.html]. Dept. of Info. and Computer Sci., Univ. of California, Irvine, CA.

....algorithm CN2 with m estimate [4, 5] and k nearest neighbour (k NN) algorithm [10] These algorithms were used both for base level and meta level learning. For base level learning, they were applied to twenty datasets from the UCI Repository of Machine Learning Databases and Domain Theories [7]. For meta level learning, the three propositional algorithms as well as two ILP systems FOIL [9] and TILDE [2] were applied to the results of base level learning, as described below. 3.1 Experimental Setting The measure of performance used in the experiments is the error rate of the classifier ....

Murphy, P. M. and Aha, D. W. (1994) UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.

....from A using the same formula as for info ML 3 Experiments 3. 1 Experimental Methodology In order to evaluate the performance of meta decision trees, we performed experiments on a collection of twenty one data sets from the UCI Repository of Machine Learning Databases and Domain Theories [11]. These data sets have been widely used in other comparative studies. Five learning algorithms were used in the base level experiments: two treelearning algorithms C4.5 [12] and LTree [8] the rule learning algorithm CN2 [4] the k nearest neighbor (k NN) algorithm [14] and a modi cation of the ....

Murphy, P. M. and Aha, D. W. (1994) UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.

....based measure with an accuracy based one. 3 Experimental comparison In order to compare the performance of meta decision trees with other combining schemes, we performed experiments on a collection of twenty one data sets from the UCI Repository of Machine Learning Databases and Domain Theories [8]. These data sets have been widely used in other comparative studies. Three learning algorithms were used in the base level experiments: tree learning algorithm J4.8, which is a re implementation of C4.5, k nearest neighbor (k NN or IBk) algorithm and naive Bayes (NB) algorithm. We used ....

Murphy, P. M. and Aha, D. W. (1994) UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.

.... performance comparison studies which confirmed the superiority of a naive Bayesian classification method over more elaborate and sophisticated IDAMAP methods [14] The experiments were usually run over the medically related benchmark datasets that reside at UCI Machine Learning repository [15]. We believe it may be the experimental design under which these data were prepared that fosters the use of naive Bayes. For instance, one of the authors recently attempted to explain the reasoning and motivation about decision trees to one of his friend physicians. After working for several ....

Murphy, P.M., and Aha, D.W., UCI Repository of machine learning databases [http://www.ics.uci.edu/mlearn/mlrepository.html]. Irvine, CA: University of California, Department of Information and Computer Science, 1994.

.... Applied Physics Laboratory, John Hopkins University, Laurel, MD 20707 and was constructed by constrained selection from a larger database held by the National Institute of Diabetes and Digestive and Kidney Diseases [44] It is publicly available from the machine learning repository at UCI [34]. All the patients represented in this dataset are females at least 21 years old of Pima Indian heritage living near Phoenix, Arizona, USA. There are eight input attributes the values of which are used to predict the output classification of testing positive for diabetes and testing negative ....

.... output variable is really a binarised form of another variable which itself is highly indicative of certain types of diabetes but does not have a one to one correspondence with the condition of being diabetic [35] To date no machine learning approach has obtained an accuracy higher than 78 [34]. The discovered ACGF models have yielded very high accuracies (79.7 ) outperforming other machine learning approaches (see Table 3) 5.3 Sin(X Y) Prediction Problem In the previous sections we have illustrated the effectiveness of Cartesian granule features in modelling classification ....

C. J. Merz and P. M. Murphy (1996) "UCI Repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA", , University of California, Irvine, CA.

.... Applied Physics Laboratory, John Hopkins University, Laurel, MD 20707 and was constructed by constrained selection from a larger database held by the National Institute of Diabetes and Digestive and Kidney Diseases [44] It is publicly available from the machine learning repository at UCI [34]. All the patients represented in this dataset are females at least 21 years old of Pima Indian heritage living near Phoenix, Arizona, USA. There are eight input attributes the values of which are used to predict the output classification of testing positive for diabetes and testing negative ....

.... output variable is really a binarised form of another variable which itself is highly indicative of certain types of diabetes but does not have a one to one correspondence with the condition of being diabetic [36] To date no machine learning approach has obtained an accuracy higher than 78 [34]. The discovered ACGF models have yielded very high accuracies (79.7 ) outperforming other machine learning approaches (see Table 3) 6. Conclusions The focus and motivation behind this work was the development of an automatic system identification process that leads to additive Cartesian ....

C. J. Merz and P. M. Murphy (1996) "UCI Repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA", University of California, Irvine, CA.

....our experiments with synthetic data, we evaluated the various approximations on real world data sets. We checked several data repositories, but could not locate data sets for discrete variable clustering. So instead, we obtained classification data sets from the UCI Machine Learning Repository (Merz and Murphy, 1996) and discarded the known class information. We used the Small Soybean (Michalski and Chilausky, 1980) Standard Audiology (Bareiss and Porter, 1987) and Lung Cancer (Hong and Yang, 1994) databases. For the Audiology data set, where both training and test data were available, we merged these data ....

Merz, C., & Murphy, P. (1996). Uci repository of machine learning databases [http://www.ics.uci.edu/ mlearn/mlrepository.html]. Tech. rep., University of California, Irvine.

....estimates can be used in CN2 for calculating the probability p(cjcond) Each of them can be also used for calculating the same term in WRAcc. 3 Experiments We performed experiments on a collection of twenty one domains from the UCI Repository of Machine Learning Databases and Domain Theories [5] and two datasets originating from mutagenesis domain [4] These domains have been widely used in other comparative studies. The domains properties (number of classes, number of examples, number of discrete, continuous and all attributes and class distribution) are given in Table 1. Table 1: ....

Murphy, P. M. and Aha, D. W. (1994) UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science. 10

....that this simple change results in a dramatic decrease of the number of learned rules, at the expense of (on average) a small drop in accuracy. 4 Experiments We performed experiments on a collection of twenty one domains from the UCI Repository of Machine Learning Databases and Domain Theories [6] and two data sets originating from mutagenesis domain [5] These domains have been widely used in other comparative studies. The domains properties (number of examples, number of discrete and continuous attributes and class distribution) are given in Table 1. Performance of the rule inducing ....

Murphy, P. M. and Aha, D. W. (1994) UCI repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.

....sample. We have not yet undertaken experiments with non random distributions, which is the ultimate target of the DAGGER algorithm. 4.1 Datasets To conduct these experiments, we were restricted to datasets with nominal attributes. We used well known datasets from the Machine Learning Repository, (Merz and Murphy, 1998). We are aware of the criticism leveled against using these datasets as the basis for comparing learning algorithms. In our defense, we wish to point out that for the first hypothesis we use them as an equality benchmark, rather than using them to prove that we do better than the weighted vote ....

C. J. Merz and P. M. Murphy (1998). UCI Repository of Machine Learning Databases [http://www.ics.uci.edu/~mlearn/MLRepository.html], University of California, Irvine.

....classification trees (nary trees on n training subsets) and with binary multi level classification trees are conducted on four datasets from the UCI machine learning repository: three MONK s problem datasets donated by Sebastian Thrun and the Tic Tac Toe Endgame dataset donated by David W. Aha [8]. The MONK s problems are a collection of three artificial binary classification problems over the same six attribute discrete domain (a1, a6) All MONK s datasets contain 432 instances without missing values, representing the full truth tables in the space of the attributes. The true ....

Merz, C.J., Murphy, P.M.: UCI Repository of Machine Learning Databases [http:// www.ics.uci.edu/ ~mlearn/ MLRepository.html]. Department of Information and Computer Science, University of California, Irvine, CA (1998)

....our results are based on a very limited comparison of symbolic and nonsymbolic data mining tools. 1 Introduction This paper presents the results of comparing symbolic and connectionist data mining methods on three different data sets from the Machine Learning Database Repository at UC Irvine [1]. We used the rule induction software CN2 [2] and simple feed forward, back propagation artificial neural networks deployed with the Stuttgart Neural Network Simulator (SNNS) package [3] The goal of this work was to analyse the performances of symbolic systems and artificial neural networks used ....

.... LaplaceAccuracy = 2 1 4 2 = 0:5 F = 2; 2) E = 3 4 5 ; 2 4 5 ) 2:4; 1:6) Significance = 2 (2 ln 2 2:4 2 ln 2 1:6 ) 0:16 The complex is better than bestcond = null, so: bestcond = hair = blonde newstar = hair = blonde Complex tested hair = red : Distribution [1 0] LaplaceAccuracy = 1 1 1 2 = 0:67 F = 1; 0) E = 3 1 5 ; 2 1 5 ) 0:6; 0:4) Significance = 2 (1 ln 1 0:6 0 ln 0 0:4 ) 1:02 The complex is better than bestcond = hair = blonde , so: bestcond = hair = red newstar = hair = blonde , hair = red The tests on the ....

C.J. Merz and P.M. Murphy (1998). UCI Repository of machine learning databases http://www.ics.uci.edu/ mlearn/MLRepository.html. Irvine, CA: University of California, Department of Information and Computer Science.

....be met in real situations if the number of missing bits in the data is relatively small, and the problems then tend to become easier. For example, many of the data sets which appear in the Machine Learning Repository of the Computer Science Department of the University of California at Irvine [30] satisfy the above condition with a small constant k. Note that, in this case, the complexity of RE(C) is polynomially equivalent to that of EXTENSION(C) This is because a pBmb ( T , F ) has a robust extension if and only if the pdBf (T # , F # ) has an extension in C, where T # (resp. F # ) ....

P. M. Murphy and D. W. Aha, UCI Repository of machine learning databases [http://www.ics.uci.edu/mlearn/MLRepository.html], University of California, Department of Information and Computer Science, 1994.

....times in order to make the cost of I O processing better visible and to give a fair account of the cost of swapping of TANE MEM with large databases. We ran the algorithms on a number of real life databases. The databases and their descriptions are available on the UCI Machine Learning Repository [13]. The number of rows, columns, and minimal dependencies found (N ) in each database are shown in Table 1. The datasets labeled Wisconsin breast cancer Theta n are concatenations of n copies of the Wisconsin breast cancer data. The set of dependencies is the same in all of them. To avoid ....

....association rules, independently observed also in [3] to find an apt generalization of both and to develop an algorithm for discovering such rules. Acknowledgements The Adult, Hepatitis, Lymphography, and Wisconsin breast cancer data have been obtained from the UCI Machine Learning Repository [13]. The Lymphography domain originates from the University Medical Centre, Institute of Oncology, Ljubljana, Yugoslavia. Thanks go to M. Zwitter and M. Soklic for providing the data. The FDEP program was obtained from I. Savnik s FDEP Home Page [16] Finally, we thank Heikki Mannila and JeanFranc ....

C. J. Merz and P. M. Murphy. UCI repository of machine learning databases [http://www.ics.uci.edu/ mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science, 1996.

....of the generalized MHR discovering algorithm. It is not appropriate to compare the holes discovered by our system with the holes produced by the algorithm in [7] because the algorithm in [7] does not allow any point in its MHRs. Our example uses the Iris data from UCI machine learning repository [8], which has 150 cases and 4 continuous attributes. Only two attributes are used here. The class information in the data is not needed. Figure 7 shows the splits produced by our modified C4.5, which form many filled and empty rectangles. The minimal size of the hole we use is 0.60.6. The shaded ....

Merz, C. J, and Murphy, P. 1996. UCI repository of machine learning database [http://www.cs.uci.edu/~mlearn/MLRepository.html].

No context found.

P.M. Murphy and D.W. Aha (1994). UCI Repository of machine learning databases [http://www.ics.uci.edu/mlearn/mlrepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.

No context found.

Murphy, P. M. and Aha, D. W. #1994# UCI repository of machine learning databases #http:##www.ics.uci.edu#~mlearn#MLRepository.html#. Irvine, CA: University of California, Department of Information and Computer Science.

No context found.

C. Merz and P. Murphy. UCI repository of machine learning databases http://www.ics.uci.edu/#mlearn/mlrepository.html, 1996. Irvine, CA: University of California, Department of Information and Computer Science.

No context found.

Merz, C.J. and P.M. Murphy, UCI Repository of machine learning databases [http://www.ics.uci.edu/~mlearn/MLRepository.html]. Irvine, CA. 1996, Irvine, CA: University of California.

No context found.

C. J. Merz, and P. Murphy. UCI repository of machine learning database [http://www.cs.uci.edu/~mlearn/MLRepository.html], 1996

No context found.

Merz, C.J., and Murphy, P.M. Uci repository of machine learning databases [http://www.ics.uci.edu/ mlearn/mlrepository.html]. Irvine, CA: University of California, Department of Information and Computer Science (1996).

No context found.

C. J. Merz and P. M. Murphy, UCI repository of machine learning databases [http://www.ics.uci.edu/#mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science, 1996.

No context found.

Merz,C.J. and Murphy,P.M., UCI repository of machine learning databases [http://www.ics.uci.edu/MLRepository.html]. Irvine, CA. University of California, Department of Information and Computer Science, 1996.

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