### Table 2. Types of Errors in Punctuation Recovery and their Causes and Potential Solutions

"... In PAGE 7: ...xception to the rule. The heuristic rules for recovering punctuations in instant messages are still evolving. Existing rules can be modified or extended to handle exceptional cases. Table2 summarizes the error types and their potential causes and ... ..."

### Table 1: Origins of user problems in WWW navigation, the proportion of users a ected, example problems, and potential solutions.

"... In PAGE 2: ... In each case, the problems are introduced and illustrated by a user scenario (in italics). Table1 further summarises the problems arising from each level, noting the proportion of users a ected and some of the potential solutions. 2.... ..."

### Table 20: Example for ? a Unit sphere and = 0, = 1 x y z potential solution X error

1994

"... In PAGE 85: ...1. Let ? = U, the unit sphere, x2 + y2 + z2 = 1 In Table20 , we choose the exact solution f1 = 1, and in Table 21, we choose f1 = 1 2 + 1 6(x2 + y2 + z2). By directly calculation from the integral equation sys- tem (5.... In PAGE 85: ...3), the two density functions and should be = 0; = 1 for Table 20, and = 1; = 0 for Table 21. So the rst term of the Laplace apos;s expansion,based on orthonormal basis, for = 1 in Table20 and = 1 in Table 21, should be p4 , we show the numerical results in Table 22. In both Table 20 and Table 21, the major errors in getting the potential solution come from calculating the Galerkin coe cients (S i; j); (Sb i; j); (K1 i; j), and (K2 i; j), Here we use: inner integration param- eter Mi = 32, and outer integration parameter Mo = 16 for calculating the above four Galerkin coe cients.... In PAGE 85: ... So the rst term of the Laplace apos;s expansion,based on orthonormal basis, for = 1 in Table 20 and = 1 in Table 21, should be p4 , we show the numerical results in Table 22. In both Table20 and Table 21, the major errors in getting the potential solution come from calculating the Galerkin coe cients (S i; j); (Sb i; j); (K1 i; j), and (K2 i; j), Here we use: inner integration param- eter Mi = 32, and outer integration parameter Mo = 16 for calculating the above four Galerkin coe cients. When calculating the right hand side of the system, note that the inner products (f1; j) and (f2; j) involve only a smooth integrand.... ..."

Cited by 1

### Table 21: Example for ? a Unit sphere and = 1, = 0 x y z potential solution X error

1994

"... In PAGE 85: ...1. Let ? = U, the unit sphere, x2 + y2 + z2 = 1 In Table 20, we choose the exact solution f1 = 1, and in Table21 , we choose f1 = 1 2 + 1 6(x2 + y2 + z2). By directly calculation from the integral equation sys- tem (5.... In PAGE 85: ... By directly calculation from the integral equation sys- tem (5.3), the two density functions and should be = 0; = 1 for Table 20, and = 1; = 0 for Table21 . So the rst term of the Laplace apos;s expansion,based on orthonormal basis, for = 1 in Table 20 and = 1 in Table 21, should be p4 , we show the numerical results in Table 22.... In PAGE 85: ...3), the two density functions and should be = 0; = 1 for Table 20, and = 1; = 0 for Table 21. So the rst term of the Laplace apos;s expansion,based on orthonormal basis, for = 1 in Table 20 and = 1 in Table21 , should be p4 , we show the numerical results in Table 22. In both Table 20 and Table 21, the major errors in getting the potential solution come from calculating the Galerkin coe cients (S i; j); (Sb i; j); (K1 i; j), and (K2 i; j), Here we use: inner integration param- eter Mi = 32, and outer integration parameter Mo = 16 for calculating the above four Galerkin coe cients.... In PAGE 85: ... So the rst term of the Laplace apos;s expansion,based on orthonormal basis, for = 1 in Table 20 and = 1 in Table 21, should be p4 , we show the numerical results in Table 22. In both Table 20 and Table21 , the major errors in getting the potential solution come from calculating the Galerkin coe cients (S i; j); (Sb i; j); (K1 i; j), and (K2 i; j), Here we use: inner integration param- eter Mi = 32, and outer integration parameter Mo = 16 for calculating the above four Galerkin coe cients. When calculating the right hand side of the system, note that the inner products (f1; j) and (f2; j) involve only a smooth integrand.... ..."

Cited by 1

### Table 1: Recognition of artifacts in cardiovascular molecular MRI. Potential solutions to these artifacts are shown in parenthesis.

"... In PAGE 4: ... In addition, these techniques work less efficiently at higher field strengths and are fairly nonlinear, particularly if the echo time is not kept extremely short [28]. Artifacts in cardiovascular molecular MRI Several important artifacts, likewise, need to be considered when using gadolinium based probes ( Table1 ). Fat has a high R1 value and appears bright on T1-weighted sequences.... In PAGE 4: ... A strong awareness of the sensitivity, specificity and artifacts produced by each imaging technique (T1, T2*, off-resonance) is thus needed (Table 1). Methods to recognize and potentially eliminate some of these possible artifacts are provided in the parentheses within Table1 . In extreme cases multi-modality imaging may be needed.... ..."

### Table 1: Origins of user problems in WWW navigation, the proportion of users a ected, example problems, and potential solutions.

1997

"... In PAGE 2: ... In each case, the problems are introduced and illustrated by a user scenario (in italics). Table1 further summarises the problems arising from each level, noting the... ..."

Cited by 6

### Table 1: 1998 OX4 POTENTIAL COLLISION SOLUTIONS Year of Impact 2014 2038 2044 2046

"... In PAGE 10: ... A shower can be decomposed into separate returns, which are continuous strings of solutions having the same close approach; they are represented by sequences of n consecutive solutions fXi; i = k; k +1; : : : ; k +(n?1)g. It is often the case that the shower at a given time contains several di erent returns [14, Table1 ]; they generally can be interpreted as primary or secondary resonant/non-resonant returns. Each individual return follows a qualitatively di erent path to get to a rendezvous with the Earth at the same time.... In PAGE 14: ...4 Virtual impactor examples Currently, the only asteroid with known collision solutions is 1998 OX4. Table1 provides details on the four 1998 OX4 collision solutions that we have identi ed. The listed values of PR assume the uniform probability density described above, i.... ..."

### Table 25: All feasible solutions for the uncertain market potential model with no promotion

2007

### Table 4 summarizes the categories of faults symptoms and some potential solutions we developed (discussed in Section 4.2.1) to reduce the vulnerability of the system to each fault symptom.

2001

"... In PAGE 7: ... Table 5 breaks down the fault symptoms for our current design. The fault symptoms are very similar to those in Table4 , but the corruption rates are reduced significantly due to the fault-tolerant measures introduced in Section 4.... In PAGE 7: ... File system error: Again, we do not attempt to fix this fault symptom because the write-through file cache is also susceptible to these errors. Note that the corruption rate for file system errors is similar between Table4 and Table 5. The slight differences are due to the nondeterminism inherent in testing a complex, timing-dependent system.... ..."

Cited by 5