### Table 4: Genetic based algorithm results. GA (40 / 35) GA (60 / 40) GA (80 / 60)

"... In PAGE 8: ... In fact, in [ 4 ] the authors refer that they had to solve them in a VAX Alpha 2100 model 300 computer. Table4 presents the results for these problems obtained in a 200 Mhz Pentium MMX PC with 16 MB RAM. In this table column \sol* quot; gives the optimal solution (underlined), when known, or the best-published solution.... ..."

### Table 1 reports the results for the three respective data collections and for the two approaches considered. The best Jm function value found and the classi cation rate are both displayed. Moreover, the average number of steps required to nd the best value is also given in the case of the genetic algorithm. As can be observed from the table, the value of the objective function generated by the genetic formulation is in most cases lower than the corresponding value from fuzzy c-means. This means that a better set of clusters is produced leading to a higher classi cation rate. It must be noted that for small numbers of clusters the results are very close for both clustering techniques. As the number of clusters increases, however, the genetic-based technique is superior. Moreover, the superiority of the genetic approach is more apparent in the case of the Thyroid database which represents the hardest problem among the three benchmarks considered. Table 1: Comparative results

1996

Cited by 5

### Table 2. Bandwidth allocations for example 1 for the sum of the objective function using.classical optimisatron techniques in comparison to genetic based optimisation techniques.

"... In PAGE 3: ... For ease of comparison the results are tabulated below. Table2 considers the sum of the objective functions and Table 3 the product of the objective function #1 Using MATLAB function constr.m for constrained non linear optimisation; uses a Sequential Quadratic Programming method.... ..."

### Table 1. Requirements for a car radio navigation system.

### Table 2: Sensors commonly used as a complement to GNSS-receivers for en- hancement of in-car navigation systems.

"... In PAGE 28: ... Further, the odometer only gives information of the traveled distance of the navigation sys- tem. Hence, except for the magnetometer, all the measurements of the sensors in Table2 only contain information on the relative movement of the vehicle and no absolute positioning or attitude information. The translation of these sensor measurements into position and attitude estimates will therefore be of an integra- tive nature requiring the initial state of the vehicle to be known, and for which the measurement errors will accumulate with time or, for the odometer, with the traveled distance.... In PAGE 81: ...06 -0.32 Axis Misalignment [ ] yz 2 zy -5 zx 3 Table2 : IMU results. The average estimate of the accelerometer cluster pa- rameters, calculated from 20 calibration of the in-house constructed IMU.... In PAGE 102: ...E13 Table2 : Setting used when evaluating the single-axis GPS-aided INS with the aid of numerical simulations. Parameter Value Parameter Value gps 0:5 [m] [Pfd;0]1;1 0:12 [m2] acc 0:01 [m=s2] [Pfd;0]2;2 0:12 [(m=s)2] gyro 0:1 [ =s] [Pfd;0]3;3 22 [( )2] T d 0:1 [s] [Pfd;0]4;4 0:12 [( =s)2] T? max 0:1 [s] [Pfd;0]5;5 ? 1=300 [s2] T? min 0 [s] [ze fd;0]6 ? 0:05[s] Ts 10 [ms] M 10 Notes: The covariance matrix Pfd;0 was initialized as a diagonal matrix with diagonal elements as given in the table.... In PAGE 102: ... 5 6. In the example, the system parameters are those given in Table2 . A Monte Carlo simulation of the single-axis navigation example was also conducted to illustrate the error model in (20) (21).... In PAGE 106: ... The expected performance of the system when evaluated for the same trajectory and with the same setting as in the previous case (see Fig. 4 and Table2 ), is given by the black (dashed) lines in Figs. 8 10.... In PAGE 108: ...as a low-pass characteristics. In Figs. 8 10, the results from the Monte Carlo simulation of the single-axis GPS-aided integrated navigation system with the feedback for the timing error are shown together with the error covariance and mean predicted from the error model in equations (52) (55). The setting used in the simulation is again given in Table2 . Fur- ther, the timing error was modeled as uniformly distributed between 0 and 100 ms and the lter was initialized for a zero time delay.... ..."

### Table 1. Parameter settings used in the experiment

"... In PAGE 9: ...25 comparing the performance of the genetic-based planning approach with and without the subgoal strategy incorporated (also called single-goal approach). Table1 shows the parameters for this experiment. Table 1.... ..."

### TABLE II SENSORS COMMONLY USED AS A COMPLEMENT TO GNSS-RECEIVERS FOR ENHANCEMENT OF IN-CAR NAVIGATION SYSTEMS.

### Table 3: Network Traffic.

1998

"... In PAGE 6: ...2 Network Traffic Another important aspect in the evaluation of Adaptive++ and in its comparison against B+ and Dynamic Aggregation is the com- munication traffic generated by TreadMarks ( Tmk ), B+, Dynamic Aggregation ( DA ), and Adaptive++ ( A++ ). Table3 presents the total amount of data and the total numberof messagestransferred on 16 processors in the presence of each prefetching technique. As one would expect, the table shows that TreadMarks trans- fers the least amount of data among the different systems in most cases.... In PAGE 7: ... Thus, Adaptive++ is the only technique that couples a serious potential for combining with a small number of useless prefetches. Table3 confirms these observations. The table shows that B+ greatly reduces the number of messages involved in standard Tread- Marks as a result of its ability to combine messages.... ..."

Cited by 21

### Table 3. Characteristics of some Genetic- based User Modeling applications. Application Input Data Outcome T I/G [23]

2004

"... In PAGE 6: ... Nevertheless, they have also been ap- plied for filtering [7]. Table3 summarizes relevant applications of GAs for UM. 2.... ..."

Cited by 2