Elio, R., Haddadi, A.: On abstract task models and conversation policies. In: Proceedings of the Agents '99 Workshop on Specifying and Implementing Conversation Policies. (1999)  Home/Search   Document Not in Database   Summary   Related Articles   Check

....its essentially contingent nature is recognised. Our solution is to model these kinds of interactions that can guide, interrupt, or redirect existing conversations by representing them as another, parallel modelling layer above that of the existing conversation layer. This idea was suggested in [2] for specific types of conversation, but we have generalised the notion and incorporated it into a Petri Net representation. Thus a conversation is a combination of protocols being instantiated and manipulated by a particular policy. 7. CONCLUSIONS We have developed an approach for managing ....

R. Elio and A. Haddadi. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89-98, May 1999.

.... agency model, and validating the implemented systems for the conformance to the semantics is also difficult [3] 6] The other approach is to use conversation policies to guide and restrict the way of communication by forcing agents to follow a predefined sequence of message exchanges [6][7][8] These two approaches are not mutually exclusive but can be used together. This paper focuses on the latter approach, and try to provide a better way of using conversation policies. Conversation policies are essential in complicated agent conversations that involve a lot of messages and ....

....are not mutually exclusive but can be used together. This paper focuses on the latter approach, and try to provide a better way of using conversation policies. Conversation policies are essential in complicated agent conversations that involve a lot of messages and possible branches [2] 6][7][8] Without using them, individual agents in each communication step may face difficulties in determining how to communicate with each other by choosing which message type, based on their own understanding and implementation of the ACL semantics. Since this approach is considered risky, it is ....

[Article contains additional citation context not shown here]

Elio, R., and Haddadi, A. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89--98, Seattle, Washington, May 1999.

....its essentially contingent nature is recognised. Our solution is to model these kinds of interactions that can guide, interrupt, or redirect existing conversations by representing them as another, parallel modelling layer above that of the existing conversation layer. This idea was suggested in [15] for specific types of conversation, but we have generalised the notion and incorporated it into a Petri Net representation. Thus a conversation is a combination of protocols being instantiated and manipulated by a particular policy. 4.1 Conversation modelling A number of modelling techniques ....

R. Elio and A. Haddadi. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89-98, May 1999.

....the same conversation. Other conversation models have been developed, using various approaches. Extended FSM models, which, like COOL, focus more on expressivity than adherence to a model, include Kuwabara et al. 21] who add inheritance to conversations, Wagner et al. 39] and Elio and Haddadi [13], who defines a multi level state machine, or Abstract Task Model (ATM) A few others have chosen to stay within the bounds of a DFA, such as Chauhan [7] who uses COOL as the basis for her multi agent development system 13 , Nodine and Unruh [32] and Pitt and Mamdani [35] who uses DFAs to ....

Renee Elio and Afsaneh Haddadi. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89--98, Seattle, Washington, May 1999.

....required to interact with a mobile agent with its own perceptions, in a changing world. Consider an operator s conversation with an autonomous mobile robot with perceptions in a environment which is constantly changing. Dialogues with such a device will be very different (see e.g. arguments in [1]) to those in the travel planning paradigm. There will be no predictable course of events in the dialogues. The device itself may need to communicate urgently with its operator. There may not be well defined endpoints to conversations, and relevant objects may appear and disappear from the ....

....due to the structure of dialogue Information States. 4. Version II: Dialogue Move Trees, Task Trees, and Automated Reasoning Many aspects of human conversational skill are missing from state of the art dialogue systems, even those which do not implement simple state transition networks (c.f. [1]) While some progress has been made in capturing structural aspects of human conversational skill, much further research and development is needed to understand the structures, knowledge, and algorithms needed to support conversation. One of our first observations has been that the adoption of ....

Renee Elio and Afsaneh Haddadi, "On abstract task models and conversation policies," in Workshop on Specifying and Implementing Conversation Policies, Autonomous Agents'99, Seattle, 1999.

....in most Computer Aided Instruction or Computer Based Training for decades) but instead should allow mixed initiatives in the course of conversations. After many years, the mixed initiative view of conversations is again considered necessary for modeling realistic Agent to Agent conversations [13,14] In Intelligent Tutoring Systems, the Computer Agent was designed to guide the User to acquire knowledge and expertise (see, for instance, the impressive achievements described in [15] Most studies on Instructional Design represent explorations concerning the pro active, autonomous behavior ....

....we could deduce that a closed world assumption was adopted in the temporal evolution of the early software Agents. It is evident that human Users evolve as a result of communicating with the outside world. Therefore, an initial plan (sequence of tasks) may be modified during the conversation [14,18]. Users do not have static plans; in fact they generate one move at a time, as a consequence of their local judgment of the state of the conversation. At least for the User, a closed world assumption is not realistic. However, this lack of attention to the User s evolving context, and evolving ....

Elio, R. and Haddadi, A. On Abstract Task Models and Conversation Policies. Autonomous Agents 99; http://www.boeing.com/special/agents99/papers.html

....its essentially contingent nature is recognised. Our solution is to model these kinds of interactions that can guide, interrupt, or redirect existing conversations by representing them as another, parallel modelling layer above that of the existing conversation layer. This idea was suggested in [7] for specific types of converation, but we have generalised the notion and incorporated it into a Petri Net representation. Thus a conversation is a combination of protocols being instantiated and manipulated by a particular policy. In Section 5 we will discuss how Petri Nets can be used on all ....

R. Elio and A. Haddadi. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89--98, May 1999.

....terms of heuristic search, similar to constraint satisfaction in that it requires the successive addition of constraints on solutions, but also analogous to game playing in that the user and system take turns. Our approach to destination advice draws heavily on an earlier analysis of the task by Elio and Haddadi (1998, 1999), which itself borrows ideas from linguistic research on speech acts (e.g. Searle, 1969) We extend upon and adapt that work as needed to conform to the requirements of speech recognition technology and the design of the user adaptive component. Our view of conversational recommendation as ....

....the user. The remaining, dialogue level, moves are required for interactions that support progress on that task. While one side of the conversation is determined by the user, the system side of the conversation is governed by a set of control rules, described in detail in Langley, Thompson, Elio and Haddadi (1999). These rules select the next operator based on the search state. The particular instantiation of that operator (for example, which attribute to ask a question about next) is selected by consulting the Retrieval Engine and conversation history. We group conversational actions into one operator if ....

Elio R. Haddadi A., `On abstract task models and conversation policies', in Proceedings of the Agents'99 Workshop on Specifying and Implementing Conversation Policies, Seattle, WA, 1999.

....terms of heuristic search, similar to constraint satisfaction in that it requires the successive addition of constraints on solutions, but also analogous to game playing in that the user and system take turns. Our approach to destination advice draws heavily on an earlier analysis of the task by Elio and Haddadi (1998, 1999), which itself borrows ideas from linguistic research on speech acts (e.g. Searle, 1969) We extend upon and adapt that work as needed to conform to the requirements of speech recognition technology and the design of the adaptive component. Our view of conversational recommendation as heuristic ....

....the user. The remaining, dialogue level, moves are required for interactions that support progress on that task. While one side of the conversation is determined by the user, the system side of the conversation is governed by a set of control rules, described in detail in Langley, Thompson, Elio and Haddadi (1999). These rules select the next operator based on the search state and the conversational context. The particular instantiation of that operator (for example, which attribute to ask a question about next) is selected by consulting the retrieval engine and conversation history. We group ....

Elio R. Haddadi A., `On abstract task models and conversation policies', in Proceedings of the Agents'99 Workshop on Specifying and Implementing Conversation Policies, Seattle, WA, 1999.

....to implement. Other conversation models have been developed, using various approaches. Extended FSM models, which, like COOL, focus more on expressivity than adherence to a model include Kuwabara et al. 22, 21] who add inheritance to conversations; Wagner et al. 46] and Elio and Haddadi [14], who defines a multilevel state machine, or ATM. A few others have chosen to stay within the bounds of a DFA, such as Chauhan [7] who uses COOL as the basis for her multi agent development system, 1 Nodine and Unruh [34, 35] who use conversation specifications to enforce correct ....

....that agent. In either case, the agent is given a new FQAN, which is derived from the Given Name of the name submitted. For example, if an agent registers orianus.local with freckles.cs[1] umbc.ans (alternatively, freckles.cs[1] umbc.http: jackal.cs. umbc.edu ans) it may receive the FQAN orianus[14].cs[1] umbc[23] ans. An alternative is to represent the name of an agent as an actual URL. While this would be enormously convenient, it creates unacceptable naming ambiguity. Any URL should be usable as a legal agent name, for reasons of flexibility, and for compatibility with systems that use ....

R. Elio and A. Haddadi. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89--98, Seattle, Washington, May 1999.

....of conversations. Other conversation models have been developed, using various approaches. Extended FSM models, which, like COOL, focus more on expressivity than adherence to a model, include Kuwabara et al. 29] who add inheritance to conversations, Wagner et al. 43] and Elio and Haddadi [13], who defines a multi level state machine, or Abstract Task Model (ATM) A few others have chosen to stay within the bounds of a DFA, such as Chauhan [6] who uses COOL as the basis for her multi agent development system 1 , Nodine and Unruh [37] and Pitt and Mamdani [40] who uses DFAs to ....

Ren'ee Elio and Afsaneh Haddadi. On abstract task models and conversation policies. In Working Notes of the Workshop on Specifying and Implementing Conversation Policies, pages 89--98, Seattle, Washington, May 1999.

....heuristics for deciding which operator to apply on each step. This treatment also has some important implications for the modeling of user preferences. In this section, we discuss each of these issues in turn. Our approach to destination advice draws heavily on an earlier analysis of the task by Elio and Haddadi (1998, 1999), which itself borrows ideas from linguistic research on speech acts (e.g. Searle, 1969) One difference between our formulations is that the previous work distinguishes between search through a task space and a dialogue space, whereas we aggregate these into search through a combined space. ....

....the dialogue operator(s) indicated by the user s most recent utterance, since these determine the system s appropriate response. 3.2 Dialogue Operators Our view of conversational recommendation as search also requires us to specify the operators that take steps through the search space. Following Elio and Haddadi (1998, 1999), we group conversational actions if they achieve the same effect, so that two superficially different utterances constitute examples of the same operator if they take the dialogue in the same direction. Table 1 summarizes the operators assumed by the Adaptive Place Advisor, which differ somewhat ....

Elio, R., & Haddadi, A. (1999). On abstract task models and conversation policies.

No context found.

Elio, R., Haddadi, A.: On abstract task models and conversation policies. In: Proceedings of the Agents '99 Workshop on Specifying and Implementing Conversation Policies. (1999)

No context found.

Renee Elio and Afsaneh Haddadi. On abstract task models and conversation policies. In Proceedings of the Agents '99 Workshop on Specifying and Implementing Conversation Policies, Seattle, Washington, USA, May 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC