### Table 1: Precision estimate

"... In PAGE 5: ... The con- straints can be either supplied manually for particular sentence by linguists, or obtained from the transformed dependency tree in PDTB. The Table1 summarizes the precision es- timates counted on real corpus data. These measurements presented here may discount the actual benefits of our approach due to the estimated 8% of mistakes in the input corpus.... ..."

### Table 3: Sensitivity and precision estimates for experimentally verified E. coli genes.

2003

"... In PAGE 12: ... The goal is to find the starts of the challenging LiD set without loosing too many of the more trivial starts of set LiC. Table3 shows that while most gene finders partially locate all genes in set LiC and LiD, there are large varia- tions in their exact localization ability. Selecting for the highest combined performance on set LiC and LiD, one sees that EasyGene, GeneMarkS and Frame-by-Frame are best.... ..."

### TABLE 4. Precision distribution estimates for selected streams

### Table 3. Estimated precision in FAQ identification

"... In PAGE 6: ...ected as containing a FAQ based on our two heuristics. Notice that 6.90% of the docu- ments in the database are not HTML pages and were discarded. Table3 shows that our simple heuristics achieved a very good precision result, since 97,57% of the pages selected were confirmed as containing a FAQ. This evaluation was made by manually inspecting each page selected by the heuristics.... ..."

### Table 9: Precision of estimator #28for increases in fraction nonconforming#29 when used with an np

"... In PAGE 17: ...he average change point estimate is 99.96. We will now consider the precision of the change point estimator. From Table9 , it can be seen that for a 50#25 increase in the process fraction nonconforming #28#0E = 1:5#29, our estimator correctly identi#0Ced the change point in 46#25 of the simulation trials. Our estimate was within 2 #285#29 subgroups of the true process change point in 82#25 #2894#25#29 of the trials.... In PAGE 17: ... Our estimate was within 2 #285#29 subgroups of the true process change point in 82#25 #2894#25#29 of the trials. It can also be seen from Table9 that for a 20#25 increase in the process fraction nonconforming #28#0E =1:2#29, our estimator correctly identi#0Ced the change point in 13#25 of the simulation runs. Our estimator was within 2 #284#29 subgroups of the true process change point in 36#25 #2850#25#29 of the... ..."

### Table 4: Precision of estimator #28for increases in fraction nonconforming#29 when used with an np

"... In PAGE 14: ...Table4 , and for decreases in the process fraction nonconforming in Table 5. From Table 4 it can be seen that the performance of the estimator improves as the magnitude of the change increases.... In PAGE 14: ...Table 5. From Table4 it can be seen that the performance of the estimator improves as the magnitude of the change increases.... ..."

### Table 5: Precision of estimator #28for decreases in fraction nonconforming#29 when used with an np

"... In PAGE 14: ...ncrease #28#0E =1:5#29 in the process fraction nonconforming, the ARL of the np chart is 6.98. For a change of magnitude #0E =1:5 the MLE estimates the process change point exactly in almost half #2849#25#29 of the simulation runs, and the estimate is within 1 subgroup of the true process change point in 73#25 of the simulation runs. The results in Table5 corroborate the previous results that have been seen and show that our proposed estimator is very good for estimating the time of a decrease in the process fraction nonconforming. For a 30#25 decrease in the process fraction nonconforming #28#0E =0:7#29 the ARL of the np chart is 313.... ..."

### Table 10: Precision of estimator #28for decreases in fraction nonconforming#29 when used with an

"... In PAGE 18: ... simulation trials. From Table10 it can be seen that the MLE has a very good performance for estimating times of decreases in the process fraction nonconforming. It can be seen that for a 30#25 decrease in the process fraction nonconforming, the estimator correctly identi#0Ced the process change point in 31#25 of the simulation trials.... In PAGE 18: ... It was within 2 #286#29 subgroups of the process change point in 67#25 #2889#25#29 of the simulation trials. It can also be seen from Table10 that for a 10#25 decrease #28#0E =0:9#29, our estimator correctly identi#0Ced the change point in 4#25 of the trials. This is a good frequency of correctly estimating the change point given that the ARL of the control chart is 536.... ..."

### Table 2: Specificity, sensitivity and precision estimates for different gene finders in E. coli.

2003

"... In PAGE 9: ...biomedcentral.com/1471-2105/4/21 Table2 shows the percentage of genes found for eight dif- the whole genome and in random sequences. The eight gene finders are: EasyGene, Glimmer2.... ..."

### Table 4: Specificity, sensitivity and precision estimates for M. tuberculosis, H. pylori J99 and B. subtilis.

2003