CiteSeerX — Table search results

Results 1 - 10 of 19,271

Table 1 Indicator Benign Malignant

in Use of a Multi-Layer Perceptron to Predict Malignancy in Ovarian Tumors

by H. Verrelst, Y. Moreau, J. Vandewalle, Ewalle J, D. Timmerman, Yves Moreau, Joos V, Dirk Timmerman

"... In PAGE 3: ... 2 Data acquisition and feature selection The data were derived from a study group of 191 consecutive patients who were referred to a single institution (University Hospitals Leuven, Belgium) from August 1994 to August 1996. Table1 lists the di erent indicators which were considered, together with their mean value and standard deviations or together with the relative presence in cases of benign and malignant tumors. Table 1 Indicator Benign Malignant... In PAGE 3: ... Every tumor was extensively examined for its morphologic characteristics. Table1 lists the selected morphologic features: presence of ab- dominal uid collection, papillary structures ( gt; 3mm), smooth internal walls, wall irregularities, whether the cysts were unilocular, multilocular-solid and/or present on both pelvic sides. All outcomes are binary valued: every observation relates to the presence (1) or absence (0) of these characteristics.... ..."

Table. Causes of Malignant and Refractory Status Epilepticus

in ORIGINAL CONTRIBUTION A “Malignant ” Variant of Status Epilepticus

by Eva Schielke, Buchheim Masuhr

Table 6 Detected Malignant Records in Wisconsin Breast Cancer Dataset

in FP-Outlier: frequent pattern based outlier detection

by Zengyou He, Xiaofei Xu, Joshua Zhexue Huang, Shengchun Deng 2002

"... In PAGE 10: ... The results of the RNN based outlier detection algorithm on this dataset were reported in [6]. Table6 shows the results produced by the 3 algorithms. Table 6 Detected Malignant Records in Wisconsin Breast Cancer Dataset ... In PAGE 10: ...8%(112) 39(100.00%) 39(100.00%) 39(100.00%) One important observation from Table6 was that, compared to the RNN algorithm, our algorithm performed the best for all cases and never the worst. 2 The resultant dataset is public available at: ... ..."

Cited by 1

Table 4. Important variables selected by FNT algorithm automatically for Malignant detection

in ENSEMBLE OF FLEXIBLE NEURAL TREES FOR BREAST CANCER DETECTION

by Yuehui Chen, Ajith Abraham, Yong Zhang, Iita Professorship Program

"... In PAGE 11: ... It should be noted that the obtained FNT classifier has smaller size and reduced features and without a significant reduce in the accuracy. The important features for constructing the FNT models are shown in Table 3 for Benign detection and Table4 for Malignant detection. In general, the ensemble of FNTs shows the best classification rate.... In PAGE 12: ... The best accuracy was o ered by the generalized ensemble method followed by the basic ensemble method. An important advantage of the FNT model is the ability to reduce the number of input variables as presented in Table 3 and Table4 . Because of this reason, it is easy to visualize the variance of the individual FNT classifiers.... ..."