### Table 1(a). Comparison of final design for ten bar truss Author Venkayya Dobbs Rizzi Khan Qian Lingxi Ringertz Schmit

2005

"... In PAGE 3: ...max(psi) 19994.6 24067.5 20361.6 24378.2 24366.7 24970.9 24415.1 24570.4 24569.8 20349.4 Table1 (b). Comparison of final design for ten bar truss (continued) Author Gellatly Schmit Tang amp; Gu Wu amp; Wang Park Ringertz Adeli Xu.... ..."

### Table 2. Comparison of final design for seventy two bar truss Author Venkayya Schmit Berke Khan Qian Lingxi Wu Jinyin Li Quanyong

2005

### Table 1. The number of correct, incorrect, and unclassified Class predictions of the logistic regression models (ClassifyR) compared to the Khan [9] and Golub [10] classification models.

"... In PAGE 4: ...igure 4. Details of the logistic regression in R. We were able to use the ClassifyR web site to generate logistic regression models for the two data sets. The overall results of the models are summarized in Table1 . The asynchronous version of ClassifyR performed as well as the other models in terms of the number of correctly classified cases.... ..."

### Table 2. Classification results for the small, round blue-cell tumor (SRBCT) data set using logistic regression with stepwise gene selection (ClassifyR) and the ANN model described by Khan [9]. The Classify-R Score is the highest numeric prediction of the four logistic regression equations. A uniform threshold of 0.5 is used for the Classify-R model. The threshold ANN Committee Vote is different for each Class.

"... In PAGE 4: ... Several of the incorrect predictions of the logistic regression model would have been unclassified if a higher threshold was used. This can be seen in Table2 and Table 3, where the individual case results of the models are shown. SRBCT ClassifyR Khan Correct 18 18 Incorrect 2 0 Unclassified 5 7 Leukemia ClassifyR Golub Correct 28 28 Incorrect 3 0 Unclassified 3 6 Table 1.... ..."

### Table 2 Number of Internet Subscribers and Users in Pakistan If Khan apos;s estimate is correct, and the percent of the population using the Internet is four times greater than the percentages shown above, then Pakistan passed from Level 1 (Embryonic, lt;.1% users) to Level 2 (Established gt;.1% users) in late 1997 at the earliest. If estimates that the number of Internet subscribers will increase by a factor of five by 2003 are correct, then it is possible that Pakistan could reach Level 3 (Common, gt;1% users) by that year. In conversations held in November, 1999, representatives of ZoooomNet stated that the number of subscribers had grown from 70,000 in 1998 to 250,000 in the fourth quarter of 1999 [47]. While it would not be surprising if these figures are inflated, they point to very rapid growth of the user base in Pakistan at present, with some possibility that Pakistan could reach Level 3 (Common) as early as the year 2000.

### Table 5: 20 genes with the largest ratios in SRBCT data set

2004

"... In PAGE 14: ... We take logarithm base 10 of the expression levels and standardize arrays before applying the classification method. Table5 shows the list of top 20 genes. Most of genes were consistently selected from the top 96 genes used for the analysis in Khan et al.... ..."

Cited by 75

### Table 1(b). Comparison of final design for ten bar truss (continued)

2005

"... In PAGE 3: ...Table1 (a). Comparison of final design for ten bar truss Author Venkayya Dobbs Rizzi Khan Qian Lingxi Ringertz Schmit Algorithm Optimality criteria Optimality criteria Optimality criteria Optimality criteria Optimality criteria Toothed method DDDU GLM Dynamic round ACCESS Reference Ref.... ..."

### Table 5 illustrates the total number of affected SLAs nAffected as a percentage of the total number of accepted SLAs (denoted as nAccepted), i.e. P = nAffected / nAccepted (in Columns 2-4). A smaller P value indicates that there were fewer active SLAs affected by failures during the H epochs.

"... In PAGE 6: ...hoja, S. Khan, K. F. Li, Optimal Admission Controllers for Service Level Agreements in Enterprise Networks, 6th SCI Conference, Orlando, FL July 14-18, 2002 Table5 , the affected SLAs in case of node failures [5].... ..."