Dialog manager

A dialog manager (DM) is a component of a dialog system (DS), responsible for the state and flow of the conversation. Usually:

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:

The human input has different meanings depending on the context. For example, in a travel-planning DS:

The meaning of the city name depends on the previously asked question. A DM may keep that question in a state variable, and use it to convert "Tel Aviv" to "I want to depart from Tel Aviv", and convert "Gaza" to "I want to arrive at Gaza".

This function is on the border between NLU and DM: in some systems it's included in the NLU, such as the context-dependent rules of Milward (2000); while in other systems it is included in the DM, such as the NP resolution module of Mirkovic and Cavedon (2005).

Another function between the NLU and DM is, determining which input utterances are part of a single utterance. Here is an example from a job negotiation dialog:

All three utterances are actually a single offer. For the second utterance, the word "and" is a clue, but for the third utterance the only possible clue is that it was said immediately after the second one. To understand this, the DM should probably keep a timestamp of each utterance.

The computer output may be made more natural, by remembering the dialog history. For example, NPCEditor (a framework for authoring characters that answer human questions) allows the author to define question-answer pairs, such that for each question there are several possible answers. The DM selects the best answer for the question, unless it was already used, in which case it selects the 2nd best answer, etc.

A similar feature exists in ChatScript (a framework for authoring chatter-bots): Each time the DS uses a certain rule, the DM marks this rule as "used", so that it won't be used again.

A recent DS for technical assistance uses advanced machine-learned rules to select the best terms for describing items. For example, if the DM notices that it's speaking with an adult, it will use terms such as "the left hand"; if it notices that it's speaking with a child, it will use less technical terms such as "the hand where you wear the clock".

...
Wikipedia