### Table 2. Global Solution Parameters in Ant Colony System

"... In PAGE 9: ... The ultimate goal of a fully decentralised system demands no global solution parameters. Table2 and Table 3 summarise the global and agent solution parameters respectively for ACS. Table 2.... ..."

### Table 3. Agent Solution Parameters for Ant Colony System

"... In PAGE 9: ... The ultimate goal of a fully decentralised system demands no global solution parameters. Table 2 and Table3 summarise the global and agent solution parameters respectively for ACS. Table 2.... ..."

### Table 11. Parameters for ant colony optimization. parameter range selected value

2006

### Table 5: Deviations Using the Tree Algorithm

2000

"... In PAGE 21: ... If the l1-norm is used only 6 TSP-tours can be constructed, one of which is Pareto optimal. Calculating the maximal deviations of these tours from Pareto optimal tours according to De nitions 1 and 2, we get the results of Table5 . Investigating Christo des apos; algorithm we see that for all three norms 4 TSP-tours can be constructed, one of which is Pareto optimal.... ..."

Cited by 9

### Table 1 Results of an ant colony for TSP composed by 10, 20 and 30 ants on 50 cities for 2500 iterations

"... In PAGE 5: ...q. (5) and the starting node for each ant is selected randomly. = ) , ( 2 0 s r n d t where n is the number of nodes in G (5) Picture 6 Initial value for (r, s) In order to test the proposed system we launched a colony composed by 10, 20 and 30 ants on a graph containing 50 cities for 2500 iterations. The results we found, shown in the Table1 , matches with the ones we expected and agrees with the results presented in [1]. Table 1 Results of an ant colony for TSP composed by 10, 20 and 30 ants on 50 cities for 2500 iterations ... ..."

### Table 1. Army ant swarm velocity.

"... In PAGE 4: ...06 m/s, 1.9 m/s respectively; Table1 ). These measurements are 3-5 times larger than those found by Swartz (1990) or Willis (1967) for statary colonies.... In PAGE 5: ... However, the usefulness of such a measure is questionable. Direction of travel ( Table1 a): Swartz (1997) reported that previous work (Franks and Fletcher, 1983) found a systematic bias in the daily direction taken by the E. Burchelli foraging column.... ..."

### Table 2: Number of errors occurring during tree induction and size of the constructed tree using the C4.5 algorithm for different combinations of features

2000

"... In PAGE 12: ... For this test the number of errors obtained during tree induction was considered as quality measure. In Table2 the experimental results are given. In the first column the employed features are listed while the second column gives the number of errors.... In PAGE 13: ...two different classes. In column 3 of Table2 , the size of the tree which reflects the complexity of the decision procedure dependent on the chosen features is shown. All numbers reported in Table 2 are based on the full training set of BGBIBK images.... In PAGE 13: ... In column 3 of Table 2, the size of the tree which reflects the complexity of the decision procedure dependent on the chosen features is shown. All numbers reported in Table2 are based on the full training set of BGBIBK images. It can be observed that no single group of features is strong enough to allow an error free tree induction.... In PAGE 13: ... The size of all possible error free decision trees varies from BFBHBL to BGBEBJ. The results shown in Table2 indicate that it is easy to construct a decision tree that can classify the given training set without any error. However, for practical applications it is more interesting to know how well the induced tree is able to generalize the training set, i.... ..."

Cited by 1

### TABLE I NOTATIONS FOR TREE CONSTRUCTION ALGORITHM

### Table 2: Parameters settings for an heterogeous colony. Parameters = (Asite,Alocal,Plocal).

"... In PAGE 12: ...he recruitment is useful on considered functions. Figure 5.b illustrate the use of heterogenous parameters. Table2 gives detailed parameters of an heterogeneous colony of 20 ants. Results encourages strongly this way.... ..."

### Table 1. Tree construction times (seconds) for 1, 2, and 4 processors

"... In PAGE 6: ...Table 1. Tree construction times (seconds) for 1, 2, and 4 processors The tree construction times in Table1 con rm the complexity of our construction algorithm. Construction times depend linearly on the scene volume and on the num- ber of processors used.... ..."