Information Technology Reference
In-Depth Information
On the other hand, a topic shift might happen gradually. Imagine the following dialog
sequence:
A:
“In which city do you live?”
B:
“Munich.”
A:
“Ah, then you are a fan of Bayern Munich?”
B:
“Actually no. I like Arsenal.”
By mentioning the concept “city”, speaker A suggests to talk about places. Speaker
B agrees to this topic by replying with an utterance containing the concept “Munich”
specifying a German city. “Munich” in turn is unrelated to the upcoming topic “Sports”,
however, it is conceptually closely connected to Munich's local soccer club “Bayern
Munich”. Thus, the dialog merges seamlessly from the topic “Places” to the topic
“Sports”. Hobbs calls this phenomenon
topic drift
[10].
2.2
Selection of Dialog Topics
Raising an issue requires choosing a dialog topic first. Thereby, the amount of possible
topics is constrained due to the given dialog scenario, the personal relation between the
dialog partners, and their cultural background. Accordingly, not every dialog topic is
appropriate for everyday small talk conversations.
Referring to Schneider [11], there are three groups of basic options for topic selec-
tion:
(1) The
immediate situation
involves all topics addressing the
frame
of the dialog
situation.
(2) The
external situation
represents the larger context of the immediate situation and
hence of its topics.
(3) The
communication situation
refers to the conversation partners and holds private
topics such as hobbies or family.
A typical small talk starts with a topic related to the immediate situation and continues
with topics from the external or communication situation. Due to these social conven-
tions, most small talk structures are very similar and ease striking up a conversation
with other, especially unknown persons.
3
Automatic Detection of Dialog Topics
Constituting a matter of course for humans, the automatic detection of dialog topics
poses a great challenge. Given a dialog situation as defined before, it has to meet sev-
eral requirements. First of all, the underlying processes have to work
online
. As dialogs
are continuous and demand adaptive moment-by-moment decisions, it is necessary to
incessantly provide the system with information about the current topic situation di-
rectly influencing the agent's conversational behavior. Additionally, this information
has to be processed within a short time frame to guarantee humanlike reaction time.
Moreover, the wide range of possible topics, for example being discussed in everyday
