### Table 1: Average AUCPR for recall 0:5 for each task using TAN as the statistical model.

2007

"... In PAGE 5: ... SAYU and SAYU-VISTA can evaluate up to 300 clauses, before a selecting a new seed or clause head. Table1 reports the average AUCPR (recall 0:5) for Cora and the p-value for a two-tailed paired t-test between SAYU- VISTA and the other two algorithms. SAYU-VISTA per- forms signi cantly better than SAYU on this domain.... In PAGE 5: ...8 1 Precision Recall SAYU-VISTA MLN SAYU Figure 2: Precision-Recall Curves comparing SAYU-VISTA, MLNs, and SAYU on UW-CSE new seed (SAYU) or a new predicate head. Table1 reports the average AUCPR for UW-CSE and the p- value for a two-tailed paired t-test comparing SAYU-VISTA to the other approaches. SAYU-VISTA comes close to per- forming signi cantly (0:05 lt; p lt; 0:06) better than SAYU on this domain.... In PAGE 6: ...Figure 3: Precision-Recall Curves comparing SAYU-VISTA, MLNs and SAYU on Mammography with an empty network structure, we begin with a network that contained both NMD features and the aggregate features. Table1 reports the average AUCPR over all folds. We use a two-tailed paired t-test to compute signi cant results, and the p-value for the test can also be found in Table 1.... In PAGE 6: ... Table 1 reports the average AUCPR over all folds. We use a two-tailed paired t-test to compute signi cant results, and the p-value for the test can also be found in Table1 . We nd no signi cant difference between SAYU-VISTA and SAYU on this task, yet SAYU-VISTA does not perform any worse than SAYU.... ..."

Cited by 4

### Table 4: Counts of the number of times the most com- monly selected metrics appear in TAN models for SLO violations in the RAMP and STEP workloads. See Ta- ble 1 for a definition of the metrics.

2004

"... In PAGE 11: ... However, it is often more effective to adapt the metrics as conditions change. To illustrate, Table4 lists the metrics selected for at least six of the SLO definitions in either the RAMP or STEP experiments. The most commonly chosen metrics differ significantly across the workloads, which stress the testbed in different ways.... ..."

Cited by 75

### Table 4: Counts of the number of times the most com- monly selected metrics appear in TAN models for SLO violations in the RAMP and STEP workloads. See Ta- ble 1 for a definition of the metrics.

2004

"... In PAGE 12: ... However, it is often more effective to adapt the metrics as conditions change. To illustrate, Table4 lists the metrics selected for at least six of the SLO definitions in either the RAMP or STEP experiments. The most commonly chosen metrics differ significantly across the workloads, which stress the testbed in different ways.... ..."

### Table 3: Predictions of models that successfully unify the gauge couplings at MX and provide enough intermediate matter to build a model of fermion masses for tan = 43.

"... In PAGE 36: ...be true, or how one could pick a model that possesses a completely infra-red stable xed point with all Yukawa couplings non-zero. However it is possible that such a model is contained in the subset of the models in Table3 which have not all been analysed in detail because of their individual algebraic complexity. The search for a completely realistic model, and the detailed comparison of its low energy predictions to data, is beyond the scope of this paper.... ..."

### Table 3: Predictions of models that successfully unify the gauge couplings at MX and provide enough intermediate matter to build a model of fermion masses for tan = 43.

"... In PAGE 36: ...be true, or how one could pick a model that possesses a completely infra-red stable xed point with all Yukawa couplings non-zero. However it is possible that such a model is contained in the subset of the models in Table3 which have not all been analysed in detail because of their individual algebraic complexity. The search for a completely realistic model, and the detailed comparison of its low energy predictions to data, is beyond the scope of this paper.... ..."

### Table 4: Predictions of models that successfully unify the gauge couplings at MX and provide enough intermediate matter to build a model of fermion masses for tan = 5.

### Table 4: Predictions of models that successfully unify the gauge couplings at MX and provide enough intermediate matter to build a model of fermion masses for tan =5.

### Table II: Fixed points for the di erent Yukawa couplings in di erent models for i) tan lt; 30 and ii) tan mt=mb. In the case of large tan , b also reaches a xed point.

### Table 1: Models in E-Banking Literature Author Endogenous Variables Exogenous Variables Tan and Teo [2000] Intention to Use Internet

"... In PAGE 2: ... His results suggest that users apos; continuance intention is determined by their satisfaction and perceived usefulness of the application. Table1 shows a number of research papers that empirically validate models in the context of e-banking. Tan and Teo [2000] found that attitudinal factors such as Internet experience, the relative advantage of online banking and perceived risk, and perceived behavioral control factors predict the intention to adopt Internet banking services.... ..."