B. V. Dasarathy, Nearest Neighbor(NN) Norms: NN Pattern Classification Techniques, IEEE Computer Society Press, 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....complete certainty for the class membership for any given measurement as will be the case for most real tasks. Somewhat surprisingly however, the optimal distance measure does satisfy the triangle inequality which is useful for some applications like efficient image retrieval [3] Most prior work [8] on the other hand have studied the use of optimal metric distance measures primarily due to strong asymptotic results for classification performance for any metric distance. Classification performance. It can be shown that the misclassification risk R (in the limit as training set size n ##) ....

B.V. Dasarathy. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques. Computer Society Press, 1991.

....unlabeled test data point x u is available before growing the decision tree. In this, they are analogous to the nearest neighbor classifier, which waits until x u is available and then bases its classification decision on the data points that are nearest to x u according to some distance measure [19, 1, 10, 25] The advantage of lazy learning is that it can focus on the neighborhood around the test point x u . In the case of probability estimates, a lazy learning algorithm can base its estimate P(y x u ) on the data points in the neighborhood of x u and thereby side step the problem that the ....

....b] indicates a uniform distribution over the [a, b] interval. P (i) represents the prior probability of class i. Cost C(i, j) C(i, i) Model i M2 Unif[0, 100] 0 M3 Unif[0, 1000] 0 M4 Unif[0, 10000] 0 P (i) P (j) 0 P (j) P (i) 0 M7 Unif[0, 10000] Unif[0, 1000] M8 Unif[0, 100] Unif[0, 10] classifier # 2 when # 1 predicts class i 1 , # 2 predicts class i 2 , and the true class is j. Given M and #, the di#erence in the costs of # 1 and # 2 can be computed by taking the dot product : M # = i 1 ,i 2 ,j M(i 1 , i 2 , j)#(i 1 , i 2 , j) BDeltaCost computes a confidence ....

B. V. Dasarathy, editor. Nearest neighbor (NN) norms: NN pattern classification techniques. IEEE Computer Society Press, Los Alamitos, CA, 1991.

....It has also been a subject of research in machine learning, statistics, pattern recognition, neura networks and other areas for several decades. As a result, many well developed approaches to it now exist, including rule induction [20, 12] decision tree induction [8, 23] instance based learning [11, 1], linear and neural classifiers [3] Bayesian learning [17, 16] and others. In classification problems, the goal is to correctly assign examples (typically described as vectors of attributes) to one of a finite number of classes. Most of the currently available algorithms for classification are ....

....Work One item for future work is to carry out experiments on additional large databases, and using other errorbased learners besides C4.5R. Of particular interest would be to apply MetaCost to algorithms that are not unstable with respect to variations in the training set, like k nearest neighbor [11] and naive Bayes [16] In its present form, MetaCost may not be very effective with these algorithms, but an alternative is readily suggested by the results of [2] and [26] Their method consists of learning multiple models using different subsets of the attributes, instead of different subsets of ....

B. W. Dasarathy, editor. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques. IEEE Computer Society Press, Los Alamitos, CA, 1991.

....used to capture single and multi modal distributions of relevant images by finding closest fitting spheres around positive examples. The second learning technique is based on k Nearest Neighhours (k NN) a technique that has been widely used in non parametric density estimation and classification [3]. A recent application to CBIR is found in the context of keyframe based video retrieval [11] The paper is structured as follows. Section 2 briefly introduces the image features used. An exposition of the learning techniques and how they can be employed for similarity computation is given in ....

B V Dasarathy, editor. Nearest Neigbour (NN) Norms: NN Pattern Classification Techniques. IEEE Computer Society Press, 1991.

....of PNN must compute the distance of the incoming feature vector from each of the many prototype feature vectors, possibly many cycles. Various methods have been found for increasing the speed of nearest neighbors classifiers, a category PNN may be considered to fall into (see, for example, [44], and [45] for a very fast tree method) The classification accuracy of fast approximations to the naive PNN may suffer at high rejection levels. For that reason, and because the naive PNN takes only a small fraction of the total time used by the PCASYS classification system (image ....

B.V. Dasarathy, editor, Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, IEEE Computer Society Press, 1991.

....abnormality; examples are shown in Fig. 14. Of course, the important question for labeled tissues is whether the label is cor rect. To evaluate the labeling, each fully labeled (i.e. normal) slice was compared with a slice segmented and labeled with the supervised k nearest neighbor algorithm [17] obtained from a study of tissue volumes [9] We found that in every case, the labels created by the unsupervised knowledge based system corresponded to the labels found on the associated seg mented region in the slice when processed by k nearest neighbors (KNN) Correspondence of labeled regions ....

B.V. Dasarthy. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques. IEEE Computer Society Press, Los Alamitos, Ca., 1991.

....research, but are not known to be the best classifiers [6, 7] In addition, there is a lack of publicly available Chinese corpus for evaluating Chinese text categorization systems. This paper reports our comparative evaluation of three machine learning methods, namely k Nearest Neighbor (kNN) [8], Support Vector Machines (SVM) 9] and Associative Resonance Associative Map (ARAM) 10] for Chinese text categorization. kNN and SVM have been reported as the top performing methods for English text categorization [7] ARAM belongs to a popularly known family of predictive selforganizing neural ....

....2 n n i , 3) where n is the number of documents in the whole training set and n i is the number of training documents in which the keyword w i appears at least once. 3. Classifiers 3.1. K Nearest Neighbor k Nearest Neighbor (kNN) is a traditional statistical pattern recognition algorithm [8]. It has been studied extensively for text categorization [7] In essence, kNN makes prediction based on the k training patterns that are closest to the unseen (testing) pattern, according to a distance metric. The commonly used distance metrics that measure the similarity between two normalized ....

B.V. Dasarathy, Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, IEEE Computer Society Press: Las Alamitos, California, 1991.

No context found.

B. V. Dasarathy, Nearest Neighbor(NN) Norms: NN Pattern Classification Techniques, IEEE Computer Society Press, 1991.

No context found.

B.V. Dasarathy (ed.). Nearest neighbour(NN) norms: NN pattern classification techniques. IEEE Computer Society Press, 1973.

No context found.

Dasarathy B.V. (Ed.), Nearest Neighbour (NN) Norms: NN Pattern Classification Techniques, IEEE Comp. Soc. Press, 1991.

No context found.

B V Dasarathy (Ed). Nearest neighbor(NN) norms: NN pattern classification techniques. IEEE Computer Society Press, 1991.

No context found.

B V Dasarathy, editor. Nearest Neighbour(NN) norms: NN Pattern Classification Techniques. IEEE Computer Society Press, 1991.

No context found.

B. V. Dasarathy. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques. McGraw-Hill Computer Science. IEEE Computer Society Press, Las Alamitos, California, 1991.

No context found.

B. V. Dasarathy, Nearest Neighbor(NN) norms: NN pattern classification techniques. IEEE Computer Society Press, 1991.

No context found.

B.V. Dasarathy, ed., Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, Los Alamitos, Calif.; IEEE CS Press, 1991.

No context found.

Dasarathy B. V. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, Los Alamitos, CA, IEEE Computer Society Press, 1991.

No context found.

Belur V. Dasarathy. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques. McGraw-Hill Computer Science Series. IEEE Computer Society Press, Las Alamitos, California, 1991.

No context found.

Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, ed., B.V. Dasarathy, IEEE Comp. Soc. Press, Los Alamitos, 1991.

No context found.

B. Dasarathy (Ed.). Nearest Neighbor (NN) norms: NN pattern classification techniques. IEEE Computer Society Press, 1991

No context found.

B. V. Dasarathy, Nearest Neighbor(NN) norms: NN pattern classification techniques. IEEE Computer Society Press, 1991.

No context found.

B.V. Dasarathy, editor. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques. IEEE Computer Society Press, Los Alamitos, California, 1991.

No context found.

B.V. Dasarathy. Nearest Neighbor(NN) Norms: NN Pattern Classification Techniques. IEEE Computer Society Press, Lost Alamitos, CA, 1990. 25, 27

No context found.

B. V. Dasarathy, Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, IEEECE1P3SH Society Press, Los Alamitos, 99 .

No context found.

B V Dasarathy (Ed). Nearest neighbor(NN) norms: NN pattern classification techniques. IEEE Computer Society Press, 1991.

No context found.

Dasarathy B. V. Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, Los Alamitos, CA, IEEE Computer Society Press, 1991.

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