Information Technology Reference
In-Depth Information
Ta b l e 4 .
Response times (in seconds) depending on the two types of difficulty
#of tree
#actions/#entities
obj. depth
1/1
1/5
5/1
5/5
1
3
0.15
0.32
0.48
1.27
2
5
0.47
0.96
1.61
3.50
3
7
2.54
4.83
7.40
13.92
4
9
18.77
34.00
39.72
68.19
5
11
153.40 267.55 154.97 276.20
three pre-selection methods. The last row shows the performance of the system includ-
ing all three enhancements. As we can see, the full combination yields an improvement
except for utterance i where the difference is negligible. The relative improvement of
the enhancements increases with the complexity of the utterances. That is to say, the
more complex the utterance, the more the speed-ups pay off.
Altogether, the complexity of the search tree is affected by the different branching
factors at each level, and the depth which depends on the number of spoken objects. The
branching factor at the first level depends on the number of actions that have the spoken
verb as a synonym. The branching factor at the second level depends on the number of
entities that have the spoken object as a synonym. At the third level the branching factor
depends on the number of parameters of the respective skill. We further evaluated our
optimised system by varying the two complexity factors independently.
Along the rows of Table 4 we varied the number of spoken objects. Along the
columns we varied the number of actions that have the spoken verb as a synonym and
the number of entities that have the spoken object as a synonym. The number of pa-
rameters of the appropriate skill are not varied, since this number already depends on
the amount of spoken objects. In this test scenario the parameters of a skill became
distinguishable for the system by providing distinct prepositions for each parameter.
Different entities became distinguishable through their attributes and the skills were
distinguishable by the number of parameters. So we had five skills with 1, 2, 3, 4 and 5
parameters, respectively.
Table 4 shows that the number of spoken objects has a greater influence on the com-
putation time than has ambiguity. This is indicated by the last two rows which only
contain measurements greater than 10 seconds. That is unacceptable for fluent human-
robot interaction. We can also observe that action pre-selection performs very well in
this test scenario. All tests in the last row address a skill with five parameters. In this test
scenario there was no other skill involving five or more parameters. As a consequence,
the action pre-selection can rule out the other four skill candidates which implies noth-
ing less than reducing the branching factor of the top node from
5
to
1
and thus reducing
the computation time by a factor of approximately 5. This also results in comparable
computation times for the combinations 1/1 (153.40 sec) and 5/1 (154.97 sec) as well
as 1/5 (267.55 sec) and 5/5 (276.20 sec).
Finally, we analysed whether the lexicon size poses a computational problem. There-
fore, we simply added 50,000 nouns to the lexicon and used the full combination test
setup from Table 3. Now, Table 5 indicates that the additional computational effort to
process the utterances with a large lexicon plays no significant role.
