### Table 3. Comparison of distributed meeting schedulers with different density of local meetings

### Table 3. RSDPOP tests on meeting scheduling.

2005

"... In PAGE 13: ...tarting meeting Mj at each time tk. Obviously, the objective is to find a schedule s.t. the overall utility is maximized. Efficiency of the solving process The results from Table3 show how our method scales up with the size of the organization, and the number of meetings to be scheduled. As expected, the overall complexity is not necessarily influenced by the sheer size of the problem to be solved, but rather by its density.... In PAGE 13: ... To our knowledge, we have solved by far the largest distributed optimization problems with a complete algorithm (second to us would be [4], with 33 agents, 12 meetings, 47 variables, 123 constraints). We also note that the number of messages recorded in Table3 is the total number of virtual variable-to-variable messages. Since in this model an agent necessarily has several internal variables, this means that the number of real agent-to-agent messages is strictly lower than that.... ..."

Cited by 2

### Table on Nominal cover to all reinforcement (including links) to meet durability requirements

### Table 1: Class Meeting Schedule

1999

"... In PAGE 2: ... When separate activities are scheduled, each course meets with its instructor in its regularly assigned classroom. Examples of common activities for all three courses include: guest lectures from industry professionals design presentations ethics presentations Table1 gives the course meeting schedule for the spring semester, 1998. Scheduling all three courses to meet at the same time allows guest lecturers from industry the opportunity to speak to all of the computer science majors.... ..."

Cited by 1

### Table 1: Class Meeting Schedule

1999

"... In PAGE 2: ... When separate activities are scheduled, each course meets with its instructor in its regularly assigned classroom. Examples of common activities for all three courses include: guest lectures from industry professionals design presentations ethics presentations Table1 gives the course meeting schedule for the spring semester, 1998. Scheduling all three courses to meet at the same time allows guest lecturers from industry the opportunity to speak to all of the computer science majors.... ..."

Cited by 1

### Table 3: Frequency of Meetings (average number of meetings per year)

1997

"... In PAGE 11: ... Mean responses by city, age and sector were also analyzed, but the results showed no statistically significant differences. Table3 shows the frequency of meetings with business associates and associations, by type of firm. The frequency of meetings averages from less than one a... ..."

### Table 6: Contention and blocking ratios of the IS and MG programs in the communi- cation/synchronization phase scheduled by the optimized local message scheduling, the self- coordinated local scheduling and the standard local scheduling.

"... In PAGE 18: ... We simulated the same heterogeneous NOW where we did contention measurement (8 Sun workstations of B+5E+2F). Table6 presents the contention ratios and blocking ratios of the simulation. The contention ratios of the two programs by simulation are consistent with our mea- surement results in Table 4.... ..."

### Table 2 Fits by Equation 2, the per-visit version of the generalized matching law, to the log ratio of run lengths and of visit durations as a function of the log composite reinforcer ratios.

"... In PAGE 7: ... The solid lines represent the best fit of Equation 2 to these data using the least squares meth- od. Table2 presents these fits, which, with the exception of run lengths for Rat 443, are good to excellent (r 2 .... In PAGE 11: ...9 . 0, for 3 rats. Within rats, there was no consistent difference in slopes, k9, for run lengths and visit durations. Table 5 also shows the slopes, a from fits by the generalized matching law ( Table2 ) and k1 from the local... In PAGE 13: ...COMPOSITE CONCURRENT PERFORMANCES Table 5 Fits by Equation 4, the local model applied to concurrent performance, to the log ratio of run lengths and of visit durations as a function of the log of the ratio of products of obtained reinforcer likelihoods. To facilitate compar- isons, a from Table2 and k1 from Table 3 are reproduced below. Rat k9 SE log m9 SE df r2 ak1 Run lengths 441 442 443 444 0.... In PAGE 13: ... It is unlikely that these local fluctuations can account for the lack of a relation among stay efficiency and the ratio of the scheduled likelihoods of re- inforcement and the direct relation between switch efficiencies and the ratio of the sched- uled likelihoods of reinforcement. DISCUSSION Composite concurrent performances, ob- tained by combining paired performances, were described by the generalized matching relation (Figures 1 and 2 and Table2 ). These paired performances were functionally unre- lated and were obtained from different ses- sions, sometimes widely separated in time.... ..."

### Table 1. Schedule evaluation features. Local Features Equivalent Boolean Features

2005

"... In PAGE 3: ... Furthermore, we wanted features that correspond directly to meeting properties or are easily computable from the calendar. Table1 summarizes the 297 features currently used (3 local features, 16 single global features, and 55 global feature combinations). The local features capture the temporal properties of a single meeting.... In PAGE 4: ... More importantly, the use of more complex kernels would reduce visibility into the learned model: the resulting user model would not explicitly specify which combinations of features are particularly important to the user. So, we instead advocate designing the feature vector to explicitly include certain feature combinations (see Table1 ) and learning simpler linear functions. 4.... ..."

Cited by 12

### Table 2: Open Meeting WWW Accesses

1995

"... In PAGE 21: ... It is also large enough for the distribution of accesses over the pages to suggest how people navigate a complex information and conversation environment. Table2 shows a nearly consistent pattern of attrition of hosts with number of transactions from the root page, with the one exception being the smaller number who accessed the Newsletters than the Comments, although access to Newsletters was closer to the root.... ..."

Cited by 14