Dialog Management System

A Dialog Management System is a management system that can solve dialog management tasks (such as dialog configurations of a dialog system).

• Context:
  • It can (typically) be a Stateful System.
  • …
• Example(s):
  • AWS Lex's Dialog Manager.
  • Nuance's Dialog Manager.
  • MindMeld's Dialog Manager.
• See: Natural Language Generation, Natural Language Understanding.

---

---

References

2017

• (Wikipedia, 2017) ⇒ https://en.wikipedia.org/wiki/Dialog_manager Retrieved:2017-4-18.
  • A dialog manager (DM) is a component of a dialog system (DS), responsible for the state and flow of the conversation. Usually:
    • The input to the DM is the human utterance, usually converted to some system-specific semantic representation by the Natural language understanding (NLU) component. For example, in a flight-planning dialog system, the input may look like "ORDER(from=TA,to=JER,date=2012-01-01)".
    • The DM usually maintains some state variables, such as the dialog history, the latest unanswered question, etc., depending on the system.
    • The output of the DM is a list of instructions to other parts of the dialog system, usually in a semantic representation, for example "TELL(flight-num=123,flight-time=12:34)". This semantic representation is usually converted to human language by the Natural language generation (NLG) component.
  • There are many different DMs that fulfill very different roles. There can even be several DM components in a single DS.

    The only thing common to all DMs is that they are stateful, in contrast to other parts of the DS (such as the NLU and NLG components), which are just stateless functions. The DM roles can roughly be divided into these groups:

    1. Input-control DMs, which enable context-dependent processing of the human utterances.
    2. Output-control DMs. which enable state-dependent generation of text.
    3. Strategic flow-control
    4. Tactic flow-control

2014

• (Griol et al., 2014) ⇒ David Griol, Zoraida Callejas, Ramón López-Cózar, and Giuseppe Riccardi. (2014). “A Domain-independent Statistical Methodology for Dialog Management in Spoken Dialog Systems.” In: Computer Speech and Language Journal, 28(3). doi:10.1016/j.csl.2013.09.002
  • QUOTE: Dialog systems (DS) allow intuitive interaction through natural language. Dialog managers are usually implemented ad hoc and difficult to adapt to new domains. A statistical methodology is proposed to reduce the effort required to develop and adapt dialog managers. User simulation is also proposed to facilitate the acquisition of the required dialog corpus. … This paper proposes a domain-independent statistical methodology to develop dialog managers for spoken dialog systems.

---