Dialogue actions for natural language interfaces
Arne Jonsson
Department of Computer and Information Science

Linkoping University, S-581 83 LINKOPING,
SWEDEN
email: arnjo@ida.liu.se
Abstract
This paper presents an action scheme for dialogue management for natural language interfaces. The scheme guides a dialogue manager
which directs the interface's dialogue with the
user, communicates with the background system, and assists the interpretation and generation modules. The dialogue manager was
designed on the basis of an investigation of
empirical material collected in Wizard of Ozexperiments. The empirical investigations revealed that in dialogues with database systems
users specify an object, or a set of objects, and
ask for domain concept information, e.g. the
value of a property of that object or set of
objects. The interface responds by performing the appropriate action, e.g. providing the
requested information or initiating a clarication subdialogue. The action to be carried out
by the interface can be determined based on
how objects and properties are specied from
information in the user utterance, the dialogue
context, and the response from the background
system and its domain model.
1 Introduction
Users of natural language interfaces should conveniently
be able to express the commands and queries that the
background system can deal with, and the system should
react quickly and accurately to all user input. Among
other things this means that the interface must be able
to cope with connected dialogue. However, it does not
mean that the interface must be able to mimic human
interaction. On the contrary, it is erroneous to assume that humans would like to interact with computers the same way as they communicate with humans (cf.
[Dahlback, 1991b; 1991a; Dahlback and Jonsson, 1992;
Dahlback et al., 1993; Krause, 1993]). Human computer
interactions have their own sublanguages (cf. [Grishman
and Kittredge, 1986]) whose characteristics often allow
a much simpler dialogue model than models capturing
human interaction.
To illustrate some properties of such human computer
interaction consider gure 1. In information retrieval
systems a common user initiative is a request for domain concept information of a specied object, or set of
objects. Utterance U11 illustrates this. The requested
domain concept information is the value of the property
shape and the domain object is the Ford Fiesta costing
26 800 crowns. Unfortunately the system could not answer the question as the property (shape) is not utilized
in the domain, instead, in utterance S12, the system provides information about its capabilities. In U13 a new
request for information on another property of the same
domain object is presented. This time the pronoun it
replaces the rephrasing of the specication of the object,
i.e. the Ford Fiesta costing 26 800 crowns. In utterance U15 the user asks for the same concept information
but related to another object, while in U17 the object
stays the same but the property is altered. In U19 the
property remains the same but this time the user utilizes a denite description to specify an object discussed
previously, and originally specied in utterance U11.
The dialogue model presented in this paper does not
intend to mimic human conversation. It is based on the
observation that for information retrieval applications a
common user initiative is a request for domain concept
information of a specied object, or set of objects (cf.
[Ahrenberg, 1987]). A dialogue manager utilizing this
information when deciding which action to perform for
user initiatives concerned with accessing the application
will provide ecient and robust user-friendly humancomputer natural language interaction.
2 The Dialogue Manager
A dialogue manager directs a natural language interface
and holds information needed by the modules in the interface, including the dialogue manager itself. The Dialogue Manager considered in this paper was designed
from an analysis of a corpus of 21 dialogues, using ve
dierent background systems [Ahrenberg et al., 1990;
Jonsson, 1991], collected in Wizard Oz-experiments
[Dahlback et al., 1993]. The Dialogue Manager need
to be customized to account for the sublanguage carried out in a specic application. Customization allows us to adapt the behaviour of the interface to the
requirements of the application (see Jonsson [1993a;
1993b] for details).
The results presented here are based on the customiza-
U11: What is the shape of Ford Fiesta costing 26 800
crowns?
S12: Wait...
Cars cannot answer questions concerning the shape of
car models.
U13: Is it rusty?
S14: Wait...
Checking...
Manufacturer Model Year Rust
Ford
Fiesta 1982
2
U15: Does the Mercedes from 1982 have any rust damage?
S16: Wait...
Checking...
Manufacturer Model Year Rust
Mercedes
200 1982
5
U17: How fast is a Mercedes 200?
S18: Wait...
Checking...
Manufacturer Model Year Rust Top Speed
Mercedes
200 1982
5
160
U19: How fast is the Fiesta?
S20: Wait...
Checking...
Manufacturer Model Year Rust Top Speed
Ford
Fiesta 1982
2
145
Figure 1: Example of human computer dialogue using
the cars system. From a corpus of dialogues collected
in Wizard of Oz-experiments. U denotes user utterances
and S utterances from the system. The corpus examples
are translated from Swedish.
tion of the dialogue manager for three applications, other
than those utilized in the design, using a set of 30 new dialogues. One of the applications, cars, allows users to
retrieve information from a consumers guide on properties of used cars. In another application, travel,
the application domain was charter trips to the Greek
archipelago. The travel application not only utilized
information retrieval but also, in one scenario, allowed
users to order a specied charter trip.
Dialogue management information is modeled in dialogue objects. These represent the constituents of the
dialogue. A dialogue object has two components. One
component contains static information describing the
properties and relations of the dialogue object. Another
is a process description of a prototypical use of the dialogue object, an action plan [Jonsson, 1991]. During
the course of interaction a dialogue tree is built up from
instances of dialogue objects (For more details on the
Dialogue Manager see Jonsson [1993a]).
Two types of static parameters are distinguished: situation parameters and content parameters. The situation
parameters specify the situation in which the dialogue
object occurs, such as Initiator, Responder and contextual information. The content parameters are needed
for focus structure and dialogue structure. Focus structure concerns the recording of entities mentioned in the
discourse to allow a user to refer to them in the course
of the interaction. Dialogue structure involves handling
the relationships between the segments in the dialogue.
U27: what hotels are there on Crete
S28: Wait...
The hotels in Platanias on Crete are:
Kronos, Agrimia, Village Suites and Villa Margarita.
U29: which one is the cheapest
S30: Wait...
Price Villa Margarita:
4/7 11/7 18/7 25/7
1 week 2725 2725 2675 2625
2 weeks 3272 3275 3225 3175
U31: what is next cheapest
S32: Wait...
Price Village Suites:
4/7 11/7 18/7 25/7
1 week 3150 3150 3100 3050
2 weeks 4025 4025 3975 3925
U33: what service can these hotels provide
S34: Wait...
The basic price at Villa Margarita includes shared
apartment.
Cleaning 3 days/week, no dish-washing. No meals.
The basic price at Village Suites includes shared
apartment.
Cleaning 3 days/week, no dish-washing. No meals.
Figure 2: Example of dialogue using the travel system.
2.1 Focus structure parameters
As discussed above, users of information retrieval systems request database information by specifying a
database object, or a set of objects, and ask for the value
of a property of that object or set of objects. The dialogue objects model database objects using a parameter termed Objects and the domain concept information
in a parameter termed Properties. The values to these
parameters depend on the background system, and the
natural language interface needs to be customized to account for the demands from each application [Jonsson,
1993b]. For the cars application a relational database
is used and the objects are cars described by the subparameters (Manufacturer, Model, Year). The travel,
application utilizes a hierarchically structured database
with the Greek archipelago on top, then the resorts and
nally the hotels at each resort. However, it turns out
that there is no need to explicitly represent the various
levels in the hierarchy. Instead one single sub-parameter
holding any of these object types is sucient. To illustrate this, consider gure 2. After utterance U27 the
value of the Objects parameter is the resort Crete. This
will be changed to a set of hotels when the response from
the background system is generated, S28.
The value to the Objects parameter can be explicitly
provided as, for instance, it is in show saab 900 of 1985
model. However, this is not often the case. Instead, the
user provides only partial information, or a new set of
objects by specifying properties, e.g. Show all medium
size cars with a safety factor larger than 4. It is also
possible to describe new objects by way of other objects,
as for example in U27 in gure 2. The Objects parameter will achieve values from such intensionally specied
object descriptions by the extensional specication provided from the database access system.
The Properties parameter models the domain concept
in a sub-parameter termed Aspect which can be specied in another sub-parameter termed Value. For instance, utterance U17 in gure 1 How fast is a Mercedes
200? provides Aspect information on the domain concept, speed which is specied by the database manager
to 160, i.e. the Value of the Aspect speed is 160.
For some applications a third focal parameter is
needed, termed Secondary Objects. Its purpose is to restrict the search in the database to allow the user to investigate objects from a subset of objects one at a time
as exemplied in gure 2. The user picks out the set of
hotels at the resort but is only interested in a subset of
them. If we apply the principle that hotels are appended
to the Objects parameter if the resort remains the same,
the Objects parameter will hold the subset requested in
U33. However, to restrict the database search in U31 to
the set specied in S28, Secondary Objects is needed to
hold the subset from which individual objects are investigated.
The focus parameters are properties of discourse segments (cf. [Zancanaro et al., 1993]), not moves. Focus is
maintained using a simple copying principle where each
new dialogue object is instantiated with a copy of the
focus parameters from the previous dialogue object (cf.
[Sene, 1992]). This forms the initial context for the dialogue object and is updated with new information from
the user initiative and the response from the background
system.
The details on how to update the focal parameters
vary and need to be considered when customizing the
dialogue objects for a specic application. For instance,
consider the system response S18 in gure 1. This response does not only contain the requested information
on the Aspect sub-parameter top speed. It also provides
information on the Aspect sub-parameter rust specied
in the previous user initiative. If the value to the Objects
parameter remains the same (or is a subset of the previous value), the value to the Properties parameter will
be the conjunction of the previous value and the new
values provided in the new move. This principle is appropriate when information is presented in tables allowing additional information to be presented conveniently
[Ahrenberg et al., 1993].
2.2 Dialogue structure parameters
The dialogue is divided into three main classes on the
basis of structural complexity. There is one class corresponding to the size of a dialogue, another class corresponding to the size of a discourse segment and a third
class corresponding to the size of a single speech act, or
dialogue move. Utterances are not analyzed as dialogue
objects, but as linguistic objects which function as vehicles of one or more moves. There are various other proposals as to the number of categories needed. They dier
mainly on the modeling of complex units that consist of
sequences of discourse segments, but do not comprise the
whole dialogue. For instance, LOKI [Wachtel, 1986] and
SUNDIAL [Bilange, 1991] use four. In LOKI the levels
are: conversation, dialogue, exchange and move. SUNDIAL uses the categories Transaction level, Exchange
level, Intervention level and Dialogue Acts. The feature
characterizing the intermediate level (i.e. the Dialogue
and Exchange levels respectively in Wachtel's and Bilange's models) is that of having a common topic, i.e.
an object whose properties are discussed over a sequence
of exchanges. However, as illustrated in gure 1, a sequence of segments may hang together in a number of
dierent ways; e.g. by being about one object for which
dierent properties are at issue. But it may also be the
other way around, so that the same property is topical, while dierent objects are talked about (cf. [Ahrenberg et al., 1990]). Thus, only one discourse segment
category is distinguished and an Initiative-response (IR)
structure is assumed (cf. adjacency-pairs [Scheglo and
Sacks, 1973]) where an initiative opens a segment by introducing a new goal and the response closes the segment
[Dahlback, 1991b].
To specify the functional role of a move we use the
parameters Type and Topic.
Type corresponds to the illocutionary type of the
move. For so-called simple service systems1 two subgoals can be identied [Hayes and Reddy, 1983, p. 266]:
1) specifying a parameter to the system and 2) obtaining
the specication of a parameter. Initiatives are categorized accordingly as being of two dierent types: 1) update, U, where users provide information to the system
and 2) question, Q, where users obtain information from
the system. Responses are categorized as answer, A, for
database answers from the system or answers to clarication requests. The Dialogue Manager utilizes other
Type categories such as Greeting, Farewell and Discourse
Continuation (DC) [Dahlback, 1991b] the latter being
used for utterances from the system whose purpose is to
keep the conversation going, but they will not be further
considered in this paper.
Topic describes which knowledge source to consult.
For information retrieval applications three dierent
knowledge sources are utilized: the database for solving
a task (T), acquiring information about the database,
system-related, (S) or, nally, the ongoing dialogue (D).
If the background system allows ordering of a specied
item a fourth category is needed to account for such utterances.
The Type/Topic parameters can be used to describe
the dialogue structure, i.e. which action to be carried
out by the interface. This in turn can be modeled in a
dialogue grammar [Jonsson, 1993a].
3 Actions for task-related initiatives
Normally a natural language interface to database information retrieval applications is user-directed, i.e. the
user initiates a request for information from the background system and the interface responds with the
requested information. The interface only takes the
initiative to begin a clarication request under three
Simple service systems \require in essence only that the
customer or client identify certain entities to the person providing the service; these entities are parameters of the service, and once they are identied the service can be provided"
[Hayes and Reddy, 1983, p. 252].
1
Objects
Properties
Correct
Correct
Partly Correct Partly Correct Aspect
Not Provided
Correct
Erroneous Value
Partly correct Ambiguous Aspect
Not provided
Incompatible
Correct
Not provided
Erroneous
Erroneous Aspect
Incompatible
(Too large to print)
Action(s)
AT
QD =AD AT
(AD )
QD =AD AT
AS
AS
AS
QD =AD AT
Table 1: A summary of the Dialogue Manager's actions
to task-related initiatives.
circumstances2 :
a diculty arises when interpreting an utterance,
e.g. unknown words or questions outside the domain
of the database.
a diculty arises when accessing the database, e.g.
when the user needs to provide a parameter for correct access.
a diculty arises in the presentation of the result
from the database access, e.g. the answer is too
large to print on one screen.
The action to be carried out for task-related questions
depends on how the information in the user initiative
together with the information copied from the previous
IR-unit and context information from the dialogue tree
and the answer from the database system specify the
values to the focal parameters Objects and Properties.
This contrasts with other structural based approaches,
such as Sitter and Stein [1992], where the user's purpose is considered primary when deciding which action
to carry out. An object or property description can be
either: correct, partly correct, incompatible, ambiguous,
erroneous, or not provided. Erroneous means that the
user has specied an object which is not in the database.
Partly correct means that the description contains at
least one correct object or property description, but also
one or more erroneous descriptions. Incompatible descriptions utilize elements which do not belong together,
e.g. Volvo Camry.
The relation between the values to the Objects and
Properties parameters and the resulting action described
in terms of Type and Topic is summarized in table 13.
Any combination of Objects and Properties in a cell in a
row results in the action to the right. From the table we
2
The system also takes the initiative to collect ordering
information.
3
When presenting the dialogue actions, Topic type will be
indicated with a subscript to the Type, e.g. AT denotes a
task-related answer. IR-units are presented as a TypeTopicpair with the Initiative separated from the Response by a
slash (/).
U17: which 10 car models between 60 000 and 70 000 crowns
are most spacious
S18: Wait...
Checking...
Information on space is either coupe or boot.
Please be more specic.
U19: best coupe
Figure 3: Example of ambiguous Aspect resulting in a
clarication request.
can identify three basic actions to task-related IR-units
depending on the values of the parameters Objects and
Properties: AT , AS , and QD =AD AT .
AT is the normal action following a QT . This describes a successful task-related user initiative followed by a successful system answer with information taken from the database. This requires correct
values for both Objects and Properties. The values
for these parameters can be taken either from the
preceding dialogue or they could be provided in the
user input. What is important is that the initiative
in context provides enough information so that it
can be used to access the background system and
that the answer from the background system is in
some sense correct. A special case is when no explicit Objects description is provided but the Properties are fully specied and can be used to access the
database, e.g. show all medium class cars costing
less than 70 000 crowns.
If the parameters Objects or Properties are partly
correct, i.e. contain one or more erroneous items,
then an answer is presented on the correctly specied items together with information about what
was erroneous, if possible.
QD =AD AT is to be considered as a special case
of the normal AT -action as specied above. This
category is concerned with cases where the system
initiates a clarication subdialogue to achieve more
information from the user in order to get fully and
correctly specied values to Objects or Properties. If
the user decides not to answer the clarication request, then the values from the initiating IR-unit are
copied to the new IR-unit and interaction proceeds
from there. The treatment of multiple sequential
clarications follows the same pattern as that for
one clarication subdialogue.
A clarication subdialogue can be initiated when
the Objects are correctly specied but the values
of the Value slot to the Properties are erroneous or
under-specied. For instance, in remove all cars
with low operational safety the expression low is too
vague. Another case is where no Aspect is provided
or the provided Aspect is ambiguous. The latter is
illustrated in utterance U17 in gure 3.
Such cases are handled by a system initiated clarication subdialogue, a QD =AD , directed from the
IR-unit which started the interaction, normally a
QT , with the under-specied or ambiguous prop-
erty copied from the initiating IR-unit. The Aspect
slot is used to hold the parameter for which the system wants an answer and the Value slot is used for
the user's answer. If the user answers correctly, as
in U19 in gure 3, the values for Properties in the
initiating IR-unit are updated. A QD =AD -unit is
identied from the type information, i.e. the Type
of the response from the user is A. Otherwise the
user move is regarded not to be an answer to the
systems clarication request. A clarication subdialogue is not initiated unless the system is able to
explicitly provide alternatives to the user.
A special case of clarication request occurs when a
correct specication of the parameters Objects and
Properties is provided, but the answer is too large
to print on the screen. In such cases the system initiates a clarication subdialogue asking the user to
restrict the number of items to be printed, for example, S2: Wait... There are 76 car models which satisfy your requirements. cars normally only shows
25 cars at a time. Do you want to see them all?.
The answer can be either a number, a restriction
such as U3: remove cars costing less than 40 000
crowns, or Yes or No. It is used to restrict the number of objects to output on the screen and also in
some cases aect the values of the Objects parameter.
AS is used for task-related user initiatives resulting in a system answer which provides information
about the database system. Information can be provided on various aspects of what type of information
there is in the database and what type of questions
that can be used to elicit this information. A typical
example is Cars cannot answer questions concerning the shape of car models. An AS is utilized for
any utterance with erroneous Objects or Aspect. Incompatible Properties and Objects also result in an
AS , this means that although both Properties and
Objects are correct, they cannot be used together.
To illustrate the action scheme consider utterance U11
What is the shape of Ford Fiesta costing 26 800 crowns?
in gure 1. This will be interpreted as a task-related
question, a QT , with correctly specied Objects parameter. However, the Aspect sub-parameter is erroneous, as
there is no information in the database on the concept
shape. Furthermore, the system can not provide alternatives to the user. Thus, the resulting action is an AS ,
S12. The next user utterance, U13, is a QT with both
correct Objects, as copied from the previous IR-unit, and
correct Aspect sub-parameter, rust. Thus, the resulting
action is an AT , S14.
It is not always possible to directly use the values in
the Objects and Properties slots, even if correctly specied. For applications such as travel, with hierarchically structured databases the Dialogue Manager sometimes needs to search the domain base or the dialogue
tree to nd an applicable object or property. For instance, if the user in the dialogue in gure 2 asks for concept information on properties associated with resorts,
such as climate, when the hotels are in focus, the domain
model is utilized to nd the appropriate resort.
There are user initiatives which do not depend on the
values of Objects and Properties, such as system-related
questions, QS , i.e. the user requests information about
the system. These are recognized on the grounds of linguistic information provided by the syntactic/semantic
analyzer [Ahrenberg, 1988].
If ordering is allowed it is important to know which
task is currently being performed, exploring the database
or ordering. This problem has been discussed by,
for instance, Ramshaw [1991], and Lambert and Carberry [1991]. They present models using three dierent,
but interacting, levels of plans to know when users stop
exploring dierent plans and instead commit themselves
to one plan. However, a result emerging from the analysis of our dialogues [Jonsson, 1993a] is that the subjects
clearly signal when they change plan, using utterances
such as I would like to order a trip for two to Lefkada.
Thus, retrieval of ordering information from the users
can be collected in a formalized fashion controlled by
the system, (cf. [Hoeppner et al., 1986]).
4 Results
Dialogue objects has been customized to meet the demands of the three systems discussed above: cars and
travel with and without ordering. The customized dialogue objects for the cars system has also been integrated with an INGRES database and interpreting modules using a grammar and a lexicon covering a subset
of the utterances found in the corpus. A context free
grammar with less than 20 rules can accurately model
the dialogue structure utilized in the corpus. The principle of copying information from one dialogue object to
the other provides the correct context for most referring
expressions. For cars only 5% required a search in the
dialogue tree. The corresponding numbers for travel
were 6% for information retrieval and 2% if ordering is
utilized (For more details on the results from customizing the dialogue and focus structures, see Jonsson [1993a]
and Ahrenberg et al. [1993]).
The action scheme presented in table 1 covers all taskrelated user initiatives utilized in the corpus. In the
cars application 85% of the user initiatives are taskrelated questions. In the travel application without ordering the number of task-related user initiatives account
for 93% of the user utterances and nally when ordering
is allowed 90% of the user utterances are task-related.
The other user initiatives are system related questions,
farewells, greetings, etc which are interpreted from linguistic information. Thus, a majority of the users' initiatives are task-related and will be handled eciently
and accurately using the action scheme.
5 Discussion
The Dialogue Manager presented in this paper is restricted to written human-computer interaction in natural language. However, when communicating with a
natural language interface, a user should not be limited
to typed keyboard input and screen output. The possibilities of using various modalities must be addressed
to further improve the interaction. Examples of sys-
tems which use a variety of modalities for both interpretation and generation include AlFresco [Stock, 1991],
XTRA [Wahlster, 1991], Voyager [Zue, 1994] and cubricon [Neal and Shapiro, 1991].
The main dierence between multi-modal interfaces
to simple service systems and conventional natural language interfaces to such applications is their ability
to utilize a combination of input and output modalities such as speech, graphics, pointing and video output. Thus, more advanced interpretation and generation
modules are required and principles for determining how
to utilize each media are needed [Arens et al., 1993].
However, the dialogue and focus structures need not
necessarily be more complicated. For instance, Voyager
[Zue, 1994] successfully utilizes the approach presented
here of copying the focus parameters from one segment to
the other [Sene, 1992]. Sitter and Stein [1992] present a
model for dialogue management to information-seeking
dialogues. The model assumes that conversation is based
on possible sequences of dialogue acts which are modeled
in a transition network. In Stein and Thiel [1993] the
model is extended to handle multi-modal interaction as
utilized in the MERIT system [Stein et al., 1992].
Thus, it seems that for simple service systems, the dialogue model presented here will be sucient not only for
natural language interfaces but also interfaces utilizing
various other modalities. However, for task-oriented dialogues, where the user's task directs the dialogue [Loo
and Bego, 1993], a model of this and the user's goals
need to be consulted in order to provide user-friendly
interaction (cf. [Burger and Marshall, 1993]). This does
not imply the necessity of a sophisticated model based
on the user's intentions. Utilizing a hierarchical structure of plans based on the various tasks possible to carry
out in the domain might do just as well (cf. [Wahlster et
al., 1993]).
6 Summary
Natural language interaction will be more robust and
habitable if the users can participate in a coherent dialogue with the system. For natural language interfaces to
information retrieval applications the necessary dialogue
actions can be determined using a straightforward solution. Users specify a database object, or set of objects,
and ask for domain concept information of that object
or objects. This is modeled in two parameters, one associated with the objects and another with the requested
properties of that object. The parameters are specied
from information in the user initiative, the discourse and
the background system and its domain model. The action to be carried out by the interface can be determined
from the specication of these objects and properties parameters.
Acknowledgments
This work results from a project on Dynamic NaturalLanguage Understanding supported by The Swedish
Council of Research in the Humanities and Social Sciences (HSFR) and The Swedish National Board for Industrial and Technical Development (NUTEK) in the
joint Research Program for Language Technology. The
work has been carried out with the members of the Natural Language Processing Laboratory at Linkoping University, Sweden, and I am especially indebted to Lars
Ahrenberg, Nils Dahlback and Ake Thuree.
References
[Ahrenberg et al., 1990] Lars Ahrenberg, Arne Jonsson,
and Nils Dahlback. Discourse representation and discourse management for natural language interfaces. In
Proceedings of the Second Nordic Conference on Text
Comprehension in Man and Machine, Taby, Sweden,
1990.
[Ahrenberg et al., 1993] Lars Ahrenberg, Arne Jonsson,
and Ake Thuree. Customizing interaction for natural language interfaces. In Workshop on Pragmatics
in Dialogue, The XIV:th Scandinavian Conference of
Linguistics and the VIII:th Conference of Nordic and
General Linguistics, Goteborg, Sweden, 1993.
[Ahrenberg, 1987] Lars Ahrenberg. Interrogative Structures of Swedish. Aspects of the Relation between
grammar and speech acts. PhD thesis, Uppsala Uni-
versity, 1987.
[Ahrenberg, 1988] Lars Ahrenberg. Functional constraints in knowledge-based natural language understanding. In Proceedings of the 12th International
Conference on Computational Linguistics, Budapest,
pages 13{18, 1988.
[Arens et al., 1993] Yigal Arens, Eduard Hovy, and
Mira Vossers. On the knowledge underlying multimedia presentations. In Mark T. Maybury, editor, Intelligent Multimedia Interfaces, pages 280{306. MITPress,
1993.
[Bilange, 1991] Eric Bilange. A task independent oral
dialogue model. In Proceedings of the Fifth Conference of the European Chapter of the Association for
Computational Linguistics, Berlin, 1991.
[Burger and Marshall, 1993] John D. Burger and Ralph
J. Marshall. The application of natural language models to intelligent multimedia. In Mark T. Maybury,
editor, Intelligent Multimedia Interfaces, pages 174 {
196. MITPress, 1993.
[Dahlback and Jonsson, 1992] Nils Dahlback and Arne
Jonsson. An empirically based computationally
tractable dialogue model. In Proceedings of the Fourteenth Annual Meeting of The Cognitive Science Society, Bloomington, Indiana, 1992.
[Dahlback et al., 1993] Nils Dahlback, Arne Jonsson,
and Lars Ahrenberg. Wizard of oz studies { why and
how. Knowledge-Based Systems, 6(4):258{266, 1993.
[Dahlback, 1991a] Nils Dahlback. Empirical analysis of
a discourse model for natural language interfaces. In
Proceedings of the Thirteenth Annual Meeting of The
Cognitive Science Society, Chicago, Illinois, pages 1{
6, 1991.
[Dahlback, 1991b] Nils Dahlback. Representations of
Discourse, Cognitive and Computational Aspects.
PhD thesis, Linkoping University, 1991.
[Grishman and Kittredge, 1986] Ralph Grishman and
Richard I. Kittredge. Analysing language in restricted
domains. Lawrence Erlbaum, 1986.
[Hayes and Reddy, 1983] Philip J. Hayes and D. Raj
Reddy. Steps toward graceful interaction in spoken
and written man-machine communication. International Journal of Man-Machine Studies, 19:231{284,
1983.
[Hoeppner et al., 1986] Wolfgang Hoeppner, Katharina
Morik, and Heinz Marburger. Talking it over: The
natural language dialog system ham-ans. In Leonard
Bolc and Matthias Jarke, editors, Cooperative Interfaces to Information Systems. Springer-Verlag, Berlin
Heidelberg, 1986.
[Jonsson, 1991] Arne Jonsson. A dialogue manager using initiative-response units and distributed control.
In Proceedings of the Fifth Conference of the European Chapter of the Association for Computational
Linguistics, Berlin, 1991.
[Jonsson, 1993a] Arne Jonsson. Dialogue Management
for Natural Language Interfaces { An Empirical Approach. PhD thesis, Linkoping University, 1993.
[Jonsson, 1993b] Arne Jonsson. A method for development of dialogue managers for natural language interfaces. In Proceedings of the Eleventh National Conference of Articial Intelligence, Washington DC, pages
190{195, 1993.
[Krause, 1993] Jurgen Krause. A multilayered empirical approach to multimodality: Towards mixed soultions of natural language and graphical interfaces. In
Mark T. Maybury, editor, Intelligent Multimedia Interfaces, pages 328 { 352. MITPress, 1993.
[Lambert and Carberry, 1991] Lynn Lambert and Sandra Carberry. A tripartite plan-based model of dialogue. In Proceedings of the 29th Annual Meeting of
the ACL, Berkeley, pages 193{200, 1991.
[Loo and Bego, 1993] W. Van Loo and H. Bego. Agent
tasks and dialogue management. In Workshop on
Pragmatics in Dialogue, The XIV:th Scandinavian
Conference of Linguistics and the VIII:th Conference
of Nordic and General Linguistics, Goteborg, Sweden,
1993.
[Neal and Shapiro, 1991] Jeannette G. Neal and Stuart C. Shapiro. Intelligent multi-media interface technology. In Joseph W. Sullivan and Sherman W.
Tyler, editors, Intelligent User Interfaces. ACM Press,
Addison-Wesley, 1991.
[Ramshaw, 1991] Lance A. Ramshaw. A three-level
model for plan exploration. In Proceedings of the 29th
Annual Meeting of the ACL, Berkeley, pages 39{46,
1991.
[Scheglo and Sacks, 1973] Emanuel A. Scheglo and
Harvey Sacks. Opening up closings. Semiotica, 7:289{
327, 1973.
[Sene, 1992] Stephanie Sene. A relaxation method
for understanding spontaneous speech utterances. In
Paper presented at the Fifth DARPA Workshop on
Speech and Natural Language, 1992.
[Sitter and Stein, 1992] Stefan Sitter and Adelheit Stein. Modeling the illocutionary aspects of informationseeking dialogues. Information Processing & Management, 28(2):165{180, 1992.
[Stein and Thiel, 1993] Adelheit Stein and Ulrich Thiel.
A conversational model of multimodal interaction in
information systems. In Proceedings of the Eleventh
National Conference of Articial Intelligence, Washington DC, pages 283 { 288, 1993.
[Stein et al., 1992] Adelheit Stein, Ulrich Thiel, and
Anne Tien. Knowledge based control of visual dialogues in information systems. In Proceedings of the
1st International Workshop on Advanced Visual Interfaces, Rome, Italy, 1992.
[Stock, 1991] Oliviero Stock. Natural language exploration of an information space: the alfresco interactive
system. In Proceedings of the Twelfth International
Joint Conference on Articial Intelligence, Sydney,
Australia, pages 972{978, 1991.
[Wachtel, 1986] Tom Wachtel. Pragmatic sensitivity in
nl interfaces and the structure of conversation. In
Proceedings of the 11th International Conference of
Computational Linguistics, University of Bonn, pages
35{42, 1986.
[Wahlster et al., 1993] Wolfgang Wahlster, Elisabeth
Andre, Wolfgang Finkler, Hans-Jurgen Protlich, and
Thomas Rist. Plan-based integration of natural language and graphics generation. Articial Intelligence,
63:387 { 427, 1993.
[Wahlster, 1991] Wolfgang Wahlster. User and discourse
models for multimodal communication. In Joseph W.
Sullivan and Sherman W. Tyler, editors, Intelligent
User Interfaces. ACM Press, Addison-Wesley, 1991.
[Zancanaro et al., 1993] Massimo Zancanaro, Oliviero
Stock, and Carlo Strapparava. Dialogue cohesion sharing and adjusting in an enhanced multimodal environment. In Proceedings of the International Joint Conference of Articial Intelligence, Chambery, France,
pages 1230{ 1236, 1993.
[Zue, 1994] Victor W. Zue. Toward systems that understand spoken language. IEEE Expert, 9:51{59, 1994.